Discussion:
When I back-up .... Coping my Entire Internal HD to an external HD
Add Reply
Daniel70
2025-03-11 10:52:17 UTC
Reply
Permalink
Some time ago, I backed up my Laptop's 500GB internal HD to an 2TB
external HD using (I forget) ... It might have been 'dd' but I doubt it.

However, the 500GB Internals Image took up 500GB on the External (i.e. a
byte-by-byte image, even the empty bytes, apparently!).

Now, when I look at that External HD using my Win 11 Desk-top, it
doesn't 'see' anything.

Is this because Linux is at a 'Higher'/'deeper' level than Windows can
'see'??

If I back-up this Win-11 Desktop to the same External HD, is there a
possibility that Win-11 will write itself over the Linux Image?? Or is
Win-11 able to detect that there is 'something' there so will go looking
for the next available UNUSED portion of the External HD??
--
Daniel70
Carlos E.R.
2025-03-11 12:16:15 UTC
Reply
Permalink
Post by Daniel70
Some time ago, I backed up my Laptop's 500GB internal HD to an 2TB
external HD using (I forget) ... It might have been 'dd' but I doubt it.
However, the 500GB Internals Image took up 500GB on the External (i.e. a
byte-by-byte image, even the empty bytes, apparently!).
Now, when I look at that External HD using my Win 11 Desk-top, it
doesn't 'see' anything.
Is this because Linux is at a 'Higher'/'deeper' level than Windows can
'see'??
Windows intentionally does not see Linux filesystems. Use Linux.
Post by Daniel70
If I back-up this Win-11 Desktop to the same External HD, is there a
possibility that Win-11 will write itself over the Linux Image?? Or is
Win-11 able to detect that there is 'something' there so will go looking
for the next available UNUSED portion of the External HD??
It is important for us to know what exactly did you use to backup your
disk and how. Without that information I can not answer your question.
--
Cheers, Carlos.
Daniel70
2025-03-11 12:34:20 UTC
Reply
Permalink
Post by Carlos E.R.
Post by Daniel70
Some time ago, I backed up my Laptop's 500GB internal HD to an 2TB
external HD using (I forget) ... It might have been 'dd' but I doubt it.
However, the 500GB Internals Image took up 500GB on the External (i.e.
a byte-by-byte image, even the empty bytes, apparently!).
Now, when I look at that External HD using my Win 11 Desk-top, it
doesn't 'see' anything.
Is this because Linux is at a 'Higher'/'deeper' level than Windows can
'see'??
Windows intentionally does not see Linux filesystems. Use Linux.
Post by Daniel70
If I back-up this Win-11 Desktop to the same External HD, is there a
possibility that Win-11 will write itself over the Linux Image?? Or is
Win-11 able to detect that there is 'something' there so will go
looking for the next available UNUSED portion of the External HD??
It is important for us to know what exactly did you use to backup your
disk and how. Without that information I can not answer your question.
It's been a while, Carlos, but, when I installed MGA-6, I would have
installed MGA-6 and then done the complete back-up then, starting from
the ISO disk.
--
Daniel70
David W. Hodgins
2025-03-11 12:21:45 UTC
Reply
Permalink
Post by Daniel70
Some time ago, I backed up my Laptop's 500GB internal HD to an 2TB
external HD using (I forget) ... It might have been 'dd' but I doubt it.
Depending on how you did it, the backup may be in a file in a partition on the 2TB drive,
or the partition table may have been copied so that the partitions on the 2TB drive are
the same as they were on the 500GB drive.
Post by Daniel70
However, the 500GB Internals Image took up 500GB on the External (i.e. a
byte-by-byte image, even the empty bytes, apparently!).
Now, when I look at that External HD using my Win 11 Desk-top, it
doesn't 'see' anything.
WIndows should see an "unknown" file system in the existing partition(s).
Post by Daniel70
Is this because Linux is at a 'Higher'/'deeper' level than Windows can
'see'??
M$ doesn't want to make it easy for windows users to use anything else.
Post by Daniel70
If I back-up this Win-11 Desktop to the same External HD, is there a
possibility that Win-11 will write itself over the Linux Image?? Or is
Win-11 able to detect that there is 'something' there so will go looking
for the next available UNUSED portion of the External HD??
It really matters how the backup was done. Windows may overwrite it or it may
allow you to create new partitions on the drive. I don't trust windows for anything,
and haven't used it much since XP. I occasionally troubleshoot things for others,
but try to avoid it as much as I can.

Don't write anything to the drive until you know exactly what is on there. Working
with low level tools like dd make it easy to wipe out data with a single typo.

Regards, Dave Hodgins
Carlos E.R.
2025-03-11 14:57:41 UTC
Reply
Permalink
Post by David W. Hodgins
Post by Daniel70
Some time ago, I backed up my Laptop's 500GB internal HD to an 2TB
external HD using (I forget) ... It might have been 'dd' but I doubt it.
Depending on how you did it, the backup may be in a file in a
partition on the 2TB drive, or the partition table may have been
copied so that the partitions on the 2TB drive are the same as they
were on the 500GB drive.
Just my thoughts.
Post by David W. Hodgins
Post by Daniel70
However, the 500GB Internals Image took up 500GB on the External (i.e. a
byte-by-byte image, even the empty bytes, apparently!).
Now, when I look at that External HD using my Win 11 Desk-top, it
doesn't 'see' anything.
WIndows should see an "unknown" file system in the existing partition(s).
And helpfully offer to format it >:-P
Post by David W. Hodgins
Post by Daniel70
Is this because Linux is at a 'Higher'/'deeper' level than Windows can
'see'??
M$ doesn't want to make it easy for windows users to use anything else.
Post by Daniel70
If I back-up this Win-11 Desktop to the same External HD, is there a
possibility that Win-11 will write itself over the Linux Image?? Or is
Win-11 able to detect that there is 'something' there so will go looking
for the next available UNUSED portion of the External HD??
It really matters how the backup was done. Windows may overwrite it
or it may allow you to create new partitions on the drive. I don't
trust windows for anything, and haven't used it much since XP. I
occasionally troubleshoot things for others, but try to avoid it as
much as I can.
Don't write anything to the drive until you know exactly what is on
there. Working with low level tools like dd make it easy to wipe out
data with a single typo.
Right. Start listing the partition table, from Linux. Then do something
like:

file -s /dev/sdZ*

It might have been 'dd', or perhaps Clonezilla. Maybe there is a single
ext4 partition there, with the image files inside.
--
Cheers, Carlos.
David W. Hodgins
2025-03-11 15:23:08 UTC
Reply
Permalink
On Tue, 11 Mar 2025 10:57:41 -0400, Carlos E.R. <***@es.invalid> wrote:
<snip>
<snip>
Post by Carlos E.R.
Post by David W. Hodgins
WIndows should see an "unknown" file system in the existing partition(s).
And helpfully offer to format it >:-P
I once experimented with putting a file system on a usb stick with mkfs.ext4 /dev/sdf
rather then creating a partition on the device like sdf1.

Made the mistake of booting windows with the usb stick plugged in. Windows doesn't
offer to format it. Windows overwrote the first track putting a partition table on the
device without any prompting to do so or message indicating it had done so.

Only later when I tried to mount the device in linux did I realize there was a problem.
Experimented some more to confirm windows was doing this.

Never let windows touch a file system that wasn't created using windows, and never let it
work with a device without a partition table, unless it's in a floppy disk drive.

Regards, Dave Hodgins
Lawrence D'Oliveiro
2025-03-12 00:31:58 UTC
Reply
Permalink
Post by David W. Hodgins
I once experimented with putting a file system on a usb stick with mkfs.ext4 /dev/sdf
rather then creating a partition on the device like sdf1.
Linux doesn’t mind about things like this. Windows does.
Java Jive
2025-03-12 13:15:48 UTC
Reply
Permalink
Post by Lawrence D'Oliveiro
Post by David W. Hodgins
I once experimented with putting a file system on a usb stick with mkfs.ext4 /dev/sdf
rather then creating a partition on the device like sdf1.
Linux doesn’t mind about things like this. Windows does.
Yes, there is something strange about Windows and USB disks, or rather I
should say sticks, as my comments below definitely applied to the USB
stick being used, but not AFAIAA to conventional HDs attached via USB.
When I was preparing my 'One Rescue USB does it all" as previously
described in this and other ngs ...

SOLVED: One Rescue USB does it all
https://news.novabbs.com/aus+uk/article-flat.php?id=1467&group=uk.comp.os.linux#1467

... I found that Windows behaved most oddly with a USB stick having a
partition table. I meant to post separately about it at the time, but
once I'd got a USB stick that worked, I forgot about this problem
encountered en voyage, so now have to rely on ageing memory. AFAICR, the
issue I encountered went like this ...

As described in the original thread, I eventually created a 256GB USB
stick with partitions as follows ...

1 8GB FAT32 Win PE + Ghost + Product Key Scanner
2 8GB FAT32 Clonezilla
3 32GB Ext4 XUbuntu 22
4 Rest NTFS Ghost Image & Installation ISO Files.

... however, on the first attempt, partitions 1 & 2 were swapped around,
and Windows didn't seem to like this at all, and wouldn't let me copy
stuff to Partition #2, then the Win PE partition. When, unwisely, I
allowed Windows to 'fix' the problems it encountered with the disk -
it said the disk needed to be formatted, and, thinking it was going to
format just Partition #2 as FAT32, I allowed it to do so, it completely
wrecked my work thus far. The partition table previously created with
GPartEd under Ubuntu 22 was completely lost, and I was left with just a
USB stick with one large 'partition' covering the entire disk. When
back in Ubuntu 22 I examined the result before starting again, it seemed
to me that Windows had formatted the USB stick as though it were a large
floppy disk, without a partition table.

Even now, when I examine the working Emergency USB stick, this time
partitioned exactly as in the table above, in the Disk Manager console
in Windows, it recognises only Partition #1 as being FAT32, giving it a
drive letter; it doesn't recognise Partition #2 as being FAT32 and
therefore doesn't give it a drive letter. I think that's why I ended up
swapping the first two partitions around to the way they are now, so
that at least I could copy stuff to the Win PE partition from within
Windows.

Que?
--
Fake news kills!

I may be contacted via the contact address given on my website:
www.macfh.co.uk
TJ
2025-03-14 15:19:35 UTC
Reply
Permalink
Post by David W. Hodgins
On Tue, 11 Mar 2025 10:57:41 -0400, Carlos E.R.
<snip>
<snip>
Post by Carlos E.R.
Post by David W. Hodgins
WIndows should see an "unknown" file system in the existing
partition(s).
And helpfully offer to format it >:-P
I once experimented with putting a file system on a usb stick with mkfs.ext4 /dev/sdf
rather then creating a partition on the device like sdf1.
Made the mistake of booting windows with the usb stick plugged in. Windows doesn't
offer to format it.  Windows overwrote the first track putting a
partition table on the
device without any prompting to do so or message indicating it had done so.
Only later when I tried to mount the device in linux did I realize there was a problem.
Experimented some more to confirm windows was doing this.
Never let windows touch a file system that wasn't created using windows, and never let it
work with a device without a partition table, unless it's in a floppy disk drive.
Regards, Dave Hodgins
Interesting. I have a few things I need Windows for but I only use it
from a VirtualBox VM, Windows 7 for the last few years, recently
"upgraded" to Windows 10. Because I have a need to access certain data
from both the host and the guest, I have one host folder that is
"shared" with the guest.

So far, Windows has made no moves to change anything but the data I've
told it to access. Could be the nature of the host/guest relationship
prevents it, I suppose.

TJ
R.Wieser
2025-03-14 15:47:28 UTC
Reply
Permalink
TJ,
Post by TJ
So far, Windows has made no moves to change anything but the data I've
told it to access.
Maybe you've forgotten how Win7 responded to USB sticks that where formatted
under XP ? Yep, every time you plugged them into Win7 wanted to "fix"
them.

Regards,
Rudy Wieser
TJ
2025-03-14 18:47:54 UTC
Reply
Permalink
Post by R.Wieser
TJ,
Post by TJ
So far, Windows has made no moves to change anything but the data I've
told it to access.
Maybe you've forgotten how Win7 responded to USB sticks that where formatted
under XP ? Yep, every time you plugged them into Win7 wanted to "fix"
them.
Regards,
Rudy Wieser
Never knew it in the first place. Last Windows I used outside of Vbox
was Win2000, many, many moons ago. I don't believe I ever formatted any
USB sticks with XP, OR Win7 from Vbox.

Never wanted to. As for digital camera memory cards from back before I
had anything to do with USB sticks, I always let the camera do that.

TJ
Paul
2025-03-15 03:39:46 UTC
Reply
Permalink
Post by R.Wieser
TJ,
Post by TJ
So far, Windows has made no moves to change anything but the data I've
told it to access.
Maybe you've forgotten how Win7 responded to USB sticks that where formatted
under XP ?   Yep, every time you plugged them into Win7 wanted to "fix"
them.
Regards,
Rudy Wieser
Never knew it in the first place. Last Windows I used outside of Vbox was Win2000, many, many moons ago. I don't believe I ever formatted any USB sticks with XP, OR Win7 from Vbox.
Never wanted to. As for digital camera memory cards from back before I had anything to do with USB sticks, I always let the camera do that.
TJ
Windows treats USB sticks with RMB=1, different than sticks
with RMB=0.

RMB is the Removable Media Bit. It causes the stick to indicate it
has removable media, in windows Disk Management.

A Windows partition may need examination, if the file system
version is different. The current NTFS is version 3.1 . But it has not
always been version 3.1 . It is stubbornly version 3.1, despite major
changes having been made to it. For example, in windows 11, you can
format a partition and use a cluster size of 1MB per cluster.
This size is not supported in older OSes (Windows 7 would not mount
such a partition). Yet the claim exists, that both OSes are "NTFS 3.1",
which is not apparent to me, when Windows 11 allows larger cluster sizes
not supported elsewhere.

EXT4 is not a "manifest constant" either. Currently, a 2024 or 2025
LiveDvD, may format a partition as EXT4 and turn on feature C12. This
partition may not be compatible with LM21.3 if such an OS is
running in the room.

While Windows partitions have a Dirty Bit, that is not the
only reason that autochk would look carefully at them. There
can be other reasons, such as a declaration of what standard
it adheres to.

Paul
R.Wieser
2025-03-15 06:50:15 UTC
Reply
Permalink
Paul,
Post by Paul
A Windows partition may need examination, if the file system
version is different.
In my case the thumbdrives are all FAT32 - somethig Windows is well
aquainted with.

As far as I know the "problem" was that there was something changed in the
non-table part of the boot record (a checksum ? an ID ? can't remember
anymore), and Win7 wanted to rewrite it.

Non-changed thumbdrives still kept working fine though.

Regards,
Rudy Wieser
Lawrence D'Oliveiro
2025-03-15 06:58:43 UTC
Reply
Permalink
Windows treats USB sticks with RMB=1, different than sticks with RMB=0.
Windows doesn’t have the idea that any storage device could be hot-
pluggable, does it.
Paul
2025-03-15 08:02:15 UTC
Reply
Permalink
Post by Lawrence D'Oliveiro
Windows treats USB sticks with RMB=1, different than sticks with RMB=0.
Windows doesn’t have the idea that any storage device could be hot-
pluggable, does it.
???

where do you get these ideas ?

In the BIOS for your computer, there are SATA ports listed.
There are enable/disable buttons (these don't affect Linux,
which is still able to detect drives set in the BIOS as disabled).
In the same interface, are "Hotplug" buttons. Turning this
on, on the port, changes the PNP information, so the
driver will be ready for the dynamic appearance of a
drive in mid-session.

One consequence of connecting a hot-plug (like an ESATA drive),
is the Safely Remove icon, adds the new drive to the removal list.
You can eject a HotPlug drive. Hopefully, this results
in the drive spinning down. You can then physically remove
the drive.

USB sticks are (obviously) HotPlug. For a period of time,
one of the Windows OSes decided to stop listing filesystems
that did not need "Safely Remove" behavior, but the changes
to that OS may have been reversed. I generally handle
all my USB sticks through that interface (do an Eject or
a Safely Remove, depending on which GUI I'm in at the time),
and it is then safe to unplug.

In windows, if you switch on HotPlug behavior on the
SATA port having the Windows C: partition on it, this
can result in the Windows C: or related materials, showing
up in the Safely Remove menu. Do not panic, as the partition
cannot be removed, since it is "busy", and attempts to
Safely Remove C: will receive a notation indicating
why it cannot be removed.

If you don't like seeing your C: in that Safely Remove
menu, then disable all the HotPlug items in the BIOS.

Paul
Lawrence D'Oliveiro
2025-03-15 22:20:57 UTC
Reply
Permalink
For a period of time, one of the Windows OSes decided to stop
listing filesystems that did not need "Safely Remove" behavior, but
the changes to that OS may have been reversed.
That’s the thing. Windows is full of assumptions as to which kinds of
devices are allowed to have which kinds of filesystems installed on them,
which kinds are to be hot-pluggable and which are not.

For example, are you allowed to put NTFS on a hot-pluggable volume?
Somehow, I don’t think so.

Linux doesn’t care, it lets you mix and match however you like.
Paul
2025-03-16 00:29:40 UTC
Reply
Permalink
Post by Lawrence D'Oliveiro
For example, are you allowed to put NTFS on a hot-pluggable volume?
Somehow, I don’t think so.
Linux doesn’t care, it lets you mix and match however you like.
Let's take a look at my Cruser Glide.

It was easily ejected, so I could move on to the next experiment.
It was used to distribute Windows Updates to three machines, so
it's been ejected and moved a number of times (two OSes times three
machines)

[Picture]

Loading Image...

*******

Now, let us try a harder one. This is a Ubuntu installer, that's been
transferred to a USB stick. The

disktype.exe /dev/sdc

--- /dev/sdc
Block device, size 58.44 GiB (62746787840 bytes)
DOS/MBR partition map
Partition 1: 58.44 GiB (62746787328 bytes, 122552319 sectors from 1)
Type 0xEE (EFI GPT protective)
GPT partition map, 248 entries
Disk size 58.44 GiB (62746787840 bytes, 122552320 sectors)
Disk GUID 912294A3-3848-DC48-A737-FA399D28FB8B
Partition 1: 5.248 GiB (5635182592 bytes, 11006216 sectors from 64)
Type Basic Data (GUID A2A0D0EB-E5B9-3344-87C0-68B6B72699C7)
Partition Name "ISO9660"
Partition GUID 912294A3-3848-DC48-A736-FA399D28FB8B
ISO9660 file system
Volume name "Ubuntu 24.04 LTS amd64"
Preparer "XORRISO-1.5.4 2021.01.30.150001, LIBISOBURN-1.5.4, LIBISOFS-1.5.4, LIBBURN-1.5.4"
Data size 5.248 GiB (5635182592 bytes, 2751554 blocks of 2 KiB)
Joliet extension, volume name "Ubuntu 24.04 LTS"
Partition 2: 4.951 MiB (5191680 bytes, 10140 sectors from 11006280)
Type EFI System (FAT) (GUID 28732AC1-1FF8-D211-BA4B-00A0C93EC93B)
Partition Name "Appended2"
Partition GUID 912294A3-3848-DC48-A735-FA399D28FB8B
FAT12 file system (hints score 4 of 5)
Volume size 4.912 MiB (5150720 bytes, 2515 clusters of 2 KiB)
Volume name "ESP"
Partition 3: 300 KiB (307200 bytes, 600 sectors from 11016420)
Type Basic Data (GUID A2A0D0EB-E5B9-3344-87C0-68B6B72699C7)
Partition Name "Gap1"
Partition GUID 912294A3-3848-DC48-A734-FA399D28FB8B
Blank disk/medium
Partition 4: 53.18 GiB (57104400384 bytes, 111532032 sectors from 11018240)
Type Unknown (GUID AF3DC60F-8384-7247-8E79-3D69D8477DE4)
Partition Name ""
Partition GUID 4991DA38-3DE6-EF44-896A-450EFF4D1AF3
Ext3 file system
Volume name "writable"
UUID 921867F9-FE74-456A-87D0-18C13AEEC93A (DCE, v4)
Last mounted at "/root/var/crash"
Volume size 53.18 GiB (57104400384 bytes, 13941504 blocks of 4 KiB)
Partition 5: unused
ISO9660 file system
Volume name "Ubuntu 24.04 LTS amd64"
Preparer "XORRISO-1.5.4 2021.01.30.150001, LIBISOBURN-1.5.4, LIBISOFS-1.5.4, LIBBURN-1.5.4"
Data size 5.253 GiB (5640747008 bytes, 2754271 blocks of 2 KiB)
El Torito boot record, catalog at 950
Bootable non-emulated image, starts at 951, preloads 2 KiB
Platform 0x00 (x86), System Type 0x00 (Empty)
Bootable non-emulated image, starts at 2751570, preloads 4.951 MiB (5191680 bytes)
Platform 0xEF (EFI), System Type 0x00 (Empty)
FAT12 file system (hints score 4 of 5)
Volume size 4.912 MiB (5150720 bytes, 2515 clusters of 2 KiB)
Volume name "ESP"
Joliet extension, volume name "Ubuntu 24.04 LTS"

Now, the Windows Disk Management view of the thing. Total screwup,
but can still be ejected. It is in the Safely Remove meu.

[Picture]

Loading Image...

Paul
Lawrence D'Oliveiro
2025-03-16 01:18:26 UTC
Reply
Permalink
Post by Paul
Now, the Windows Disk Management view of the thing. Total screwup,
but can still be ejected. It is in the Safely Remove meu.
Would you trust it to “safely remove” something it cannot even understand
properly in the first place?
Paul
2025-03-16 02:44:54 UTC
Reply
Permalink
Post by Lawrence D'Oliveiro
Post by Paul
Now, the Windows Disk Management view of the thing. Total screwup,
but can still be ejected. It is in the Safely Remove meu.
Would you trust it to “safely remove” something it cannot even understand
properly in the first place?
Yes, because anything with an erroneous or weird determination,
does not get mounted. The partitions are not being listed in
the upper portion of Disk Management, and that area is kinda the
equivalent of /etc/mtab for listing things.

Since the device isn't really mounted (that I could determine!),
using the Safely Remove is more ceremony than with a purpose.
But part of trying it, of using the control, is to see if the OS
is in a weird state with respect to the device.

Disk Management also has an Offline control for storage devices.
And that can be used, if you discover a bug in something, and
need a second method for "making a storage device safe".

Paul
Lawrence D'Oliveiro
2025-03-16 06:33:26 UTC
Reply
Permalink
Post by Paul
Since the device isn't really mounted (that I could determine!),
With Windows, you can never be sure ...
Dan Purgert
2025-03-17 08:57:11 UTC
Reply
Permalink
Post by Lawrence D'Oliveiro
For a period of time, one of the Windows OSes decided to stop
listing filesystems that did not need "Safely Remove" behavior, but
the changes to that OS may have been reversed.
That’s the thing. Windows is full of assumptions as to which kinds of
devices are allowed to have which kinds of filesystems installed on them,
which kinds are to be hot-pluggable and which are not.
For example, are you allowed to put NTFS on a hot-pluggable volume?
Somehow, I don’t think so.
Practically every USB stick in existence would like to disagree.
--
|_|O|_|
|_|_|O| Github: https://github.com/dpurgert
|O|O|O| PGP: DDAB 23FB 19FA 7D85 1CC1 E067 6D65 70E5 4CE7 2860
Anssi Saari
2025-03-17 14:05:15 UTC
Reply
Permalink
Post by Dan Purgert
Post by Lawrence D'Oliveiro
For example, are you allowed to put NTFS on a hot-pluggable volume?
Somehow, I don’t think so.
Practically every USB stick in existence would like to disagree.
Those are usually FAT formatted. FAT32 or the newer exFAT. No issue
reformatting to NTFS though.

I've actually found NTFS on a USB SSD to be surprisingly widely
supported on media players and TVs and such. I've used it on Android
too. So NTFS has become my go-to portable FS.
Carlos E.R.
2025-03-17 18:23:03 UTC
Reply
Permalink
Post by Anssi Saari
Post by Dan Purgert
Post by Lawrence D'Oliveiro
For example, are you allowed to put NTFS on a hot-pluggable volume?
Somehow, I don’t think so.
Practically every USB stick in existence would like to disagree.
Those are usually FAT formatted. FAT32 or the newer exFAT. No issue
reformatting to NTFS though.
I've actually found NTFS on a USB SSD to be surprisingly widely
supported on media players and TVs and such. I've used it on Android
too. So NTFS has become my go-to portable FS.
exFAT is better, but few TV sets support it, while many support NTFS.
--
Cheers, Carlos.
Paul
2025-03-17 19:21:27 UTC
Reply
Permalink
Post by Carlos E.R.
Post by Anssi Saari
Post by Dan Purgert
Post by Lawrence D'Oliveiro
For example, are you allowed to put NTFS on a hot-pluggable volume?
Somehow, I don’t think so.
Practically every USB stick in existence would like to disagree.
Those are usually FAT formatted. FAT32 or the newer exFAT. No issue
reformatting to NTFS though.
I've actually found NTFS on a USB SSD to be surprisingly widely
supported on media players and TVs and such. I've used it on Android
too. So NTFS has become my go-to portable FS.
exFAT is better, but few TV sets support it, while many support NTFS.
EXFAT requires a license.

NTFS requires a license.

FAT32 requires a license (or at least TomTom may have discovered it did :-) )

One difference is, some applications of EXFAT are
covered by FRAND (where EXFAT is defined in a standards doc,
as the default filesystem for a certain piece of hardware).

In any case, whether license terms are fair and reasonable,
or are the normal kind of license terms, equipment makers
will not pay a penny more for licensing. It does not latter
what the license fee is, they don't want to pay it, whatever it is.

NTFS is journaled. NTFS is slightly harder on a USB stick,
from a wear perspective. You can improve the characteristics
a tiny bit, by disabling "LastAccessed".

FAT32 and EXFAT are not journaled. You can set the cluster
size on FAT32 and EXFAT (make it greater than or equal to
the flash page size).

BTFS supports extra large clusters, but the option is not
backward compatible (a Win11 NTFS partition with one megabyte
clusters, cannot be mounted by Windows XP, or by Windows 7).
And that doesn't include testing the Linux response, as if
an option isn't intended to be compatible, nobody is going to use it.

Paul
Carlos E. R.
2025-03-17 21:04:12 UTC
Reply
Permalink
Post by Paul
Post by Carlos E.R.
Post by Anssi Saari
Post by Dan Purgert
Post by Lawrence D'Oliveiro
For example, are you allowed to put NTFS on a hot-pluggable volume?
Somehow, I don’t think so.
Practically every USB stick in existence would like to disagree.
Those are usually FAT formatted. FAT32 or the newer exFAT. No issue
reformatting to NTFS though.
I've actually found NTFS on a USB SSD to be surprisingly widely
supported on media players and TVs and such. I've used it on Android
too. So NTFS has become my go-to portable FS.
exFAT is better, but few TV sets support it, while many support NTFS.
EXFAT requires a license.
NTFS requires a license.
FAT32 requires a license (or at least TomTom may have discovered it did :-) )
I remember that, there was a court case.
Post by Paul
One difference is, some applications of EXFAT are
covered by FRAND (where EXFAT is defined in a standards doc,
as the default filesystem for a certain piece of hardware).
In any case, whether license terms are fair and reasonable,
or are the normal kind of license terms, equipment makers
will not pay a penny more for licensing. It does not latter
what the license fee is, they don't want to pay it, whatever it is.
Ah, I understand.
Post by Paul
NTFS is journaled. NTFS is slightly harder on a USB stick,
from a wear perspective. You can improve the characteristics
a tiny bit, by disabling "LastAccessed".
I did not know that.
Post by Paul
FAT32 and EXFAT are not journaled. You can set the cluster
size on FAT32 and EXFAT (make it greater than or equal to
the flash page size).
Similar to disabling the journal on ext4. And there is no licensing on it.
Post by Paul
BTFS
BTFS?
Post by Paul
supports extra large clusters, but the option is not
backward compatible (a Win11 NTFS partition with one megabyte
clusters, cannot be mounted by Windows XP, or by Windows 7).
And that doesn't include testing the Linux response, as if
an option isn't intended to be compatible, nobody is going to use it.
Paul
Right.
--
Cheers,
Carlos E.R.
Lawrence D'Oliveiro
2025-03-17 23:19:43 UTC
Reply
Permalink
BTFS
BTFS?
That’s the thing, Linux is so versatile, one of its supported filesystems
is actually a K-Pop group.
Anssi Saari
2025-03-19 14:45:54 UTC
Reply
Permalink
Post by Carlos E.R.
exFAT is better, but few TV sets support it, while many support NTFS.
So how is exFAT better then? As a portable FS?
Ant
2025-03-19 16:05:06 UTC
Reply
Permalink
Post by Anssi Saari
Post by Carlos E.R.
exFAT is better, but few TV sets support it, while many support NTFS.
So how is exFAT better then? As a portable FS?
exFAT can handle bigger files and partitions. FAT32 can't. See
https://duckduckgo.com/?q=exfat+vs.+fat32 for the details. Also, exFAT
is better for portabilities.
--
"The lips of the righteous nourish many, but fools die for lack of judgment." --Proverbs 10:21. Bruce Willis is 70, March Madness, & Doyers won 2 regular games in JP!
Note: A fixed width font (Courier, Monospace, etc.) is required to see this signature correctly.
/\___/\ Ant(Dude) @ http://aqfl.net & http://antfarm.home.dhs.org.
/ /\ /\ \ Please nuke ANT if replying by e-mail.
| |o o| |
\ _ /
( )
Lawrence D'Oliveiro
2025-03-19 21:00:05 UTC
Reply
Permalink
Post by Ant
exFAT can handle bigger files and partitions.
But it doesn’t offer the option for journalling to guard against
filesystem corruption on crashes or improper removal/shutdown, does it.
Paul
2025-03-20 07:04:27 UTC
Reply
Permalink
Post by Lawrence D'Oliveiro
Post by Ant
exFAT can handle bigger files and partitions.
But it doesn’t offer the option for journalling to guard against
filesystem corruption on crashes or improper removal/shutdown, does it.
I bet most of the SD storage devices you've got in the
house, were abused at one time or another, and how often
do those get corrupted ? I don't have a lot of SDs, but
I've been popping the one I've got in and out of equipment
for years. Never once had an issue. FAT32. 32GB size.

I think journaling is an excellent feature on a HDD or
on an SSD (wear leveling). Journaling is a less good choice
on an SD or on a USB flash stick (because the storage PHY
isn't as sophisticated, and it is easier to burn a hole
in those). If we could have flash sticks with SLC or MLC,
my opinion might change then, but TLC or QLC devices
are pretty sloppy pieces of crap. And those need static
and dynamic wear leveling to give a good lifespan.

I don't think I have any ExFAT partitions here at the moment.
Usually, anything prepared that way, got paved over, through
no fault of their own. Other file systems end up on the devices
instead.

Paul
Lawrence D'Oliveiro
2025-03-20 22:02:32 UTC
Reply
Permalink
If we could have flash sticks with SLC or MLC, my opinion might
change then, but TLC or QLC devices are pretty sloppy pieces of
crap. And those need static and dynamic wear leveling to give a good
lifespan.
Speaking of which, Linux offers one or two filesystems which are
specifically designed for use on such media, with wear-levelling built
into the storage allocation algorithms themselves.
Carlos E.R.
2025-03-21 10:50:11 UTC
Reply
Permalink
Post by Lawrence D'Oliveiro
If we could have flash sticks with SLC or MLC, my opinion might
change then, but TLC or QLC devices are pretty sloppy pieces of
crap. And those need static and dynamic wear leveling to give a good
lifespan.
Speaking of which, Linux offers one or two filesystems which are
specifically designed for use on such media, with wear-levelling built
into the storage allocation algorithms themselves.
I remember one that was very promising, a four letter name that I have
forgotten; but the support was removed in my distro (openSUSE) because
the specs changed on every release of the distro, and a stick formatted
one year would be unreadable the next year or something of the sort.

It was industry designed for sticks and cards.
--
Cheers, Carlos.
Carlos E.R.
2025-03-21 11:21:22 UTC
Reply
Permalink
Post by Carlos E.R.
Post by Lawrence D'Oliveiro
If we could have flash sticks with SLC or MLC, my opinion might
change then, but TLC or QLC devices are pretty sloppy pieces of
crap. And those need static and dynamic wear leveling to give a good
lifespan.
Speaking of which, Linux offers one or two filesystems which are
specifically designed for use on such media, with wear-levelling built
into the storage allocation algorithms themselves.
I remember one that was very promising, a four letter name that I have
forgotten; but the support was removed in my distro (openSUSE) because
the specs changed on every release of the distro, and a stick formatted
one year would be unreadable the next year or something of the sort.
It was industry designed for sticks and cards.
f2fs
--
Cheers, Carlos.
Lawrence D'Oliveiro
2025-03-24 00:50:58 UTC
Reply
Permalink
Post by Carlos E.R.
Post by Lawrence D'Oliveiro
Speaking of which, Linux offers one or two filesystems which are
specifically designed for use on such media, with wear-levelling built
into the storage allocation algorithms themselves.
I remember one that was very promising, a four letter name that I have
forgotten; but the support was removed in my distro (openSUSE) because
the specs changed on every release of the distro, and a stick formatted
one year would be unreadable the next year or something of the sort.
It was industry designed for sticks and cards.
f2fs
I see it is supported in my current Debian Unstable 6.12.19 kernel. A
quick search did not turn up any reports of version-incompatibility
issues. Though the Arch Wiki <https://wiki.archlinux.org/title/F2FS> has a
comprehensive list of usage tips as well as a pretty detailed version
history.

I see nilfs2 mentioned as another similar contender.
Carlos E.R.
2025-03-24 13:23:00 UTC
Reply
Permalink
Post by Lawrence D'Oliveiro
Post by Carlos E.R.
Post by Lawrence D'Oliveiro
Speaking of which, Linux offers one or two filesystems which are
specifically designed for use on such media, with wear-levelling built
into the storage allocation algorithms themselves.
I remember one that was very promising, a four letter name that I have
forgotten; but the support was removed in my distro (openSUSE) because
the specs changed on every release of the distro, and a stick formatted
one year would be unreadable the next year or something of the sort.
It was industry designed for sticks and cards.
f2fs
I see it is supported in my current Debian Unstable 6.12.19 kernel. A
quick search did not turn up any reports of version-incompatibility
issues. Though the Arch Wiki <https://wiki.archlinux.org/title/F2FS> has a
comprehensive list of usage tips as well as a pretty detailed version
history.
I see nilfs2 mentioned as another similar contender.
Archived-At:
<https://lists.opensuse.org/archives/list/***@lists.opensuse.org/message/RJAFG564TR3ZYQSAJVJDSJPHDNPKPR35/>


<https://wiki.archlinux.org/title/F2FS>

Warning: The data contained on F2FS partitions can become unusable if
the kernel version on the running machine is older than the kernel
version used to create the partition. For example, this limitation can
manifest if the F2FS partition was created on a mainline kernel provided
by linux yet the system has a need to downgrade to an older series of
kernels provided by linux-lts. See FS#69363.



<https://www.reddit.com/r/openSUSE/comments/enlu5b/speaking_of_lack_of_f2fs_support_by_tumbleweed/>



rbrownsuse • 5y ago
Profile Badge for the Achievement Top 1% Commenter Top 1% Commenter

I'm not sure I like the idea of a booting to a filesystem that is known
to have very serious flaws, including no protection from causing the
entire system from crashing if the kernel tries to mount a F2FS
filesystem created in an older version of the kernel.

Not only is that downright terrifying in general use (just wait till
those Debian systems patch to a new version of F2FS..) but it's also a
wonderful denial of service vector..use all the fancy USB or network
hotplugability you can do with storage devices these days and watch as
the kernel tries to mount invalid old F2FS filesystems and panic all
systems....

Yeah..thanks but no thanks.. openSUSE is better off with F2FS disabled
by default

(See https://bugzilla.opensuse.org/show_bug.cgi?id=1109665 before
arguing with this comment)

...

xeq937

openSUSE doesn't support F2FS, unless you recompile the kernel manually,
as far as I know. You might try exFAT with your phone and openSUSE instead.


xeq937 • 5y ago

You can either recompile openSUSE kernel (which is a tremendous pain and
fraught with pitfalls), or use another distro. I'd go with the latter.
Unfortunately openSUSE has some sort of vendetta against F2FS.

...


mvribeiro • 5y ago

Thank you!

I think it will be compiled with the m option in the future, so that the
module gets compiled so it works after disabling the blacklist. Let's
see if it's there in the next kernel update.

Edit:
https://bugzilla.opensuse.org/show_bug.cgi?id=1173546

...


mvribeiro • 5y ago • Edited 5y ago

No! It's still blacklisted.

But after the next minor kernel update it may be possible to mount F2FS
after simply removing the blacklist.

This unreleased binary kernel rpm already has it:

https://build.opensuse.org/package/binaries/Kernel:HEAD/kernel-default/standard

I checked the contents of that x86_64.rpm and it does have a f2fs.ko.xz.

I won't install it yet because I need nvidia for two screens, but
shortly it'll probably be available on TW.

Edit: This brown guy is worse than Linus. People shouldn't let him bark
around out of his leash tarnishing the name of openSUSE devs. It's gross.






Things got easier later. Archived-At:
<https://lists.opensuse.org/archives/list/***@lists.opensuse.org/message/NYSLWKRMM65QY44GRG2RBOHND7EBAO5O/>

Huh?

***@tw:~> /usr/sbin/modinfo f2fs
filename: /usr/lib/modules/6.11.0-1-default/kernel/fs/f2fs/f2fs.ko.zst
softdep: pre: crc32
license: GPL
description: Flash Friendly File System
...
suserelease: openSUSE Tumbleweed
...
signer: openSUSE Secure Boot CA
--
Cheers, Carlos.
Carlos E.R.
2025-03-20 11:24:07 UTC
Reply
Permalink
Post by Lawrence D'Oliveiro
Post by Ant
exFAT can handle bigger files and partitions.
But it doesn’t offer the option for journalling to guard against
filesystem corruption on crashes or improper removal/shutdown, does it.
Perfect.

You do not want journalling on an usb stick or memory card.
--
Cheers, Carlos.
Lawrence D'Oliveiro
2025-03-20 22:01:02 UTC
Reply
Permalink
Post by Carlos E.R.
Post by Lawrence D'Oliveiro
Post by Ant
exFAT can handle bigger files and partitions.
But it doesn’t offer the option for journalling to guard against
filesystem corruption on crashes or improper removal/shutdown, does it.
Perfect.
You do not want journalling on an usb stick or memory card.
But SSDs are also built on flash memory technology; do you disable
journalling on those as well?
Carlos E.R.
2025-03-21 10:51:45 UTC
Reply
Permalink
Post by Lawrence D'Oliveiro
Post by Carlos E.R.
Post by Lawrence D'Oliveiro
Post by Ant
exFAT can handle bigger files and partitions.
But it doesn’t offer the option for journalling to guard against
filesystem corruption on crashes or improper removal/shutdown, does it.
Perfect.
You do not want journalling on an usb stick or memory card.
But SSDs are also built on flash memory technology; do you disable
journalling on those as well?
No, they have wear levelling, and an expected lifetime with normal usage
patterns that is quite long.
--
Cheers, Carlos.
Paul
2025-03-21 14:16:08 UTC
Reply
Permalink
Post by Lawrence D'Oliveiro
Post by Carlos E.R.
Post by Lawrence D'Oliveiro
Post by Ant
exFAT can handle bigger files and partitions.
But it doesn’t offer the option for journalling to guard against
filesystem corruption on crashes or improper removal/shutdown, does it.
Perfect.
You do not want journalling on an usb stick or memory card.
But SSDs are also built on flash memory technology; do you disable
journalling on those as well?
No, they have wear levelling, and an expected lifetime with normal usage patterns that is quite long.
Exactly. SSDs algorithm and processing power (I read of an
SSD yesterday with a five core ARM processor in it), ensures
that the entire wear life of the device (number of cells times cycles)
is harvested. USB sticks don't even come remotely close to that. Some
USB sticks, don't even seem to follow what technical information
is available for them. Either their flash chips are entire crap
(should have been thrown out at flash factory), or, something
is very wrong with the controller.

Paul
Carlos E.R.
2025-03-22 12:52:52 UTC
Reply
Permalink
Post by Paul
Post by Lawrence D'Oliveiro
Post by Carlos E.R.
Post by Lawrence D'Oliveiro
Post by Ant
exFAT can handle bigger files and partitions.
But it doesn’t offer the option for journalling to guard against
filesystem corruption on crashes or improper removal/shutdown, does it.
Perfect.
You do not want journalling on an usb stick or memory card.
But SSDs are also built on flash memory technology; do you disable
journalling on those as well?
No, they have wear levelling, and an expected lifetime with normal usage patterns that is quite long.
Exactly. SSDs algorithm and processing power (I read of an
SSD yesterday with a five core ARM processor in it), ensures
that the entire wear life of the device (number of cells times cycles)
is harvested. USB sticks don't even come remotely close to that. Some
USB sticks, don't even seem to follow what technical information
is available for them. Either their flash chips are entire crap
(should have been thrown out at flash factory), or, something
is very wrong with the controller.
I just realized I have an nvme with 72713 hours of use. Probably the first one I bought.


=== START OF INFORMATION SECTION ===
Model Family: SandForce Driven SSDs
Device Model: KINGSTON SMS200S3120G
Serial Number: ...
LU WWN Device Id: 5 0026b7 26901494e
Firmware Version: 608ABBF0
User Capacity: 120,034,123,776 bytes [120 GB]
Sector Size: 512 bytes logical/physical
Rotation Rate: Solid State Device
TRIM Command: Available
Device is: In smartctl database 7.3/5528
ATA Version is: ATA8-ACS, ACS-2 T13/2015-D revision 3
SATA Version is: SATA 3.0, 6.0 Gb/s (current: 6.0 Gb/s)
Local Time is: Sat Mar 22 13:14:02 2025 CET
SMART support is: Available - device has SMART capability.
SMART support is: Enabled

...

SMART Attributes Data Structure revision number: 10
Vendor Specific SMART Attributes with Thresholds:
ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE
1 Raw_Read_Error_Rate 0x0032 095 095 050 Old_age Always - 0/38481593
5 Retired_Block_Count 0x0033 100 100 003 Pre-fail Always - 0
9 Power_On_Hours_and_Msec 0x0032 017 017 000 Old_age Always - 72713h+43m+19.000s
12 Power_Cycle_Count 0x0032 100 100 000 Old_age Always - 184
171 Program_Fail_Count 0x000a 100 100 000 Old_age Always - 0
172 Erase_Fail_Count 0x0032 100 100 000 Old_age Always - 0
174 Unexpect_Power_Loss_Ct 0x0030 000 000 000 Old_age Offline - 134
177 Wear_Range_Delta 0x0000 000 000 000 Old_age Offline - 1
181 Program_Fail_Count 0x000a 100 100 000 Old_age Always - 0
182 Erase_Fail_Count 0x0032 100 100 000 Old_age Always - 0
187 Reported_Uncorrect 0x0012 100 100 000 Old_age Always - 0
189 Airflow_Temperature_Cel 0x0000 045 113 000 Old_age Offline - 45 (Min/Max 0/113)
194 Temperature_Celsius 0x0022 045 113 000 Old_age Always - 45 (Min/Max 0/113)
195 ECC_Uncorr_Error_Count 0x001c 120 120 000 Old_age Offline - 0/38481593
196 Reallocated_Event_Count 0x0033 100 100 003 Pre-fail Always - 0
201 Unc_Soft_Read_Err_Rate 0x001c 120 120 000 Old_age Offline - 0/38481593
204 Soft_ECC_Correct_Rate 0x001c 120 120 000 Old_age Offline - 0/38481593
230 Life_Curve_Status 0x0013 100 100 000 Pre-fail Always - 100
231 SSD_Life_Left 0x0000 094 094 011 Old_age Offline - 34359738368
233 SandForce_Internal 0x0032 000 000 000 Old_age Always - 40546
234 SandForce_Internal 0x0032 000 000 000 Old_age Always - 14524
241 Lifetime_Writes_GiB 0x0032 000 000 000 Old_age Always - 14524
242 Lifetime_Reads_GiB 0x0032 000 000 000 Old_age Always - 8232
244 Unknown_Attribute 0x0000 090 090 010 Old_age Offline - 20906303


I just run a short test, but it doesn't show - or they count hours differently:

SMART Self-test log structure revision number 1
Num Test_Description Status Remaining LifeTime(hours) LBA_of_first_error
# 1 Short offline Completed without error 00% 7178 -
# 2 Extended offline Completed without error 00% 7168 -
# 3 Short offline Completed without error 00% 7166 -
--
Cheers, Carlos.
Paul
2025-03-22 18:18:34 UTC
Reply
Permalink
Post by Carlos E.R.
Post by Paul
Post by Lawrence D'Oliveiro
Post by Carlos E.R.
Post by Lawrence D'Oliveiro
Post by Ant
exFAT can handle bigger files and partitions.
But it doesn’t offer the option for journalling to guard against
filesystem corruption on crashes or improper removal/shutdown, does it.
Perfect.
You do not want journalling on an usb stick or memory card.
But SSDs are also built on flash memory technology; do you disable
journalling on those as well?
No, they have wear levelling, and an expected lifetime with normal usage patterns that is quite long.
Exactly. SSDs algorithm and processing power (I read of an
SSD yesterday with a five core ARM processor in it), ensures
that the entire wear life of the device (number of cells times cycles)
is harvested. USB sticks don't even come remotely close to that. Some
USB sticks, don't even seem to follow what technical information
is available for them. Either their flash chips are entire crap
(should have been thrown out at flash factory), or, something
is very wrong with the controller.
I just realized I have an nvme with 72713 hours of use. Probably the first one I bought.
=== START OF INFORMATION SECTION ===
Model Family:     SandForce Driven SSDs
Device Model:     KINGSTON SMS200S3120G
Serial Number:    ...
LU WWN Device Id: 5 0026b7 26901494e
Firmware Version: 608ABBF0
User Capacity:    120,034,123,776 bytes [120 GB]
Sector Size:      512 bytes logical/physical
Rotation Rate:    Solid State Device
TRIM Command:     Available
Device is:        In smartctl database 7.3/5528
ATA Version is:   ATA8-ACS, ACS-2 T13/2015-D revision 3
SATA Version is:  SATA 3.0, 6.0 Gb/s (current: 6.0 Gb/s)
Local Time is:    Sat Mar 22 13:14:02 2025 CET
SMART support is: Available - device has SMART capability.
SMART support is: Enabled
...
SMART Attributes Data Structure revision number: 10
ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
  1 Raw_Read_Error_Rate     0x0032   095   095   050    Old_age   Always       -       0/38481593
  5 Retired_Block_Count     0x0033   100   100   003    Pre-fail  Always       -       0
  9 Power_On_Hours_and_Msec 0x0032   017   017   000    Old_age   Always       -       72713h+43m+19.000s
 12 Power_Cycle_Count       0x0032   100   100   000    Old_age   Always       -       184
171 Program_Fail_Count      0x000a   100   100   000    Old_age   Always       -       0
172 Erase_Fail_Count        0x0032   100   100   000    Old_age   Always       -       0
174 Unexpect_Power_Loss_Ct  0x0030   000   000   000    Old_age   Offline      -       134
177 Wear_Range_Delta        0x0000   000   000   000    Old_age   Offline      -       1
181 Program_Fail_Count      0x000a   100   100   000    Old_age   Always       -       0
182 Erase_Fail_Count        0x0032   100   100   000    Old_age   Always       -       0
187 Reported_Uncorrect      0x0012   100   100   000    Old_age   Always       -       0
189 Airflow_Temperature_Cel 0x0000   045   113   000    Old_age   Offline      -       45 (Min/Max 0/113)
194 Temperature_Celsius     0x0022   045   113   000    Old_age   Always       -       45 (Min/Max 0/113)
195 ECC_Uncorr_Error_Count  0x001c   120   120   000    Old_age   Offline      -       0/38481593
196 Reallocated_Event_Count 0x0033   100   100   003    Pre-fail  Always       -       0
201 Unc_Soft_Read_Err_Rate  0x001c   120   120   000    Old_age   Offline      -       0/38481593
204 Soft_ECC_Correct_Rate   0x001c   120   120   000    Old_age   Offline      -       0/38481593
230 Life_Curve_Status       0x0013   100   100   000    Pre-fail  Always       -       100
231 SSD_Life_Left           0x0000   094   094   011    Old_age   Offline      -       34359738368
233 SandForce_Internal      0x0032   000   000   000    Old_age   Always       -       40546
234 SandForce_Internal      0x0032   000   000   000    Old_age   Always       -       14524
241 Lifetime_Writes_GiB     0x0032   000   000   000    Old_age   Always       -       14524
242 Lifetime_Reads_GiB      0x0032   000   000   000    Old_age   Always       -       8232
244 Unknown_Attribute       0x0000   090   090   010    Old_age   Offline      -       20906303
SMART Self-test log structure revision number 1
Num  Test_Description    Status                  Remaining  LifeTime(hours)  LBA_of_first_error
# 1  Short offline       Completed without error       00%      7178         -
# 2  Extended offline    Completed without error       00%      7168         -
# 3  Short offline       Completed without error       00%      7166         -
That's amazing, that a 120GB drive is still alive. Some of those
die due to firmware issues.

it could be a SATA type NVME, rather than a PCIe.

The entry in /dev should help you identify what it is listed under.

As far as I know, Sandforce did compressing controllers for SATA,
and Kingston was their major customer. I could not tell you
whether Sandforce was still in business or not.

Paul
Carlos E.R.
2025-03-22 22:20:33 UTC
Reply
Permalink
Post by Paul
Post by Carlos E.R.
Post by Paul
Post by Lawrence D'Oliveiro
Post by Carlos E.R.
Post by Lawrence D'Oliveiro
Post by Ant
exFAT can handle bigger files and partitions.
But it doesn’t offer the option for journalling to guard against
filesystem corruption on crashes or improper removal/shutdown, does it.
Perfect.
You do not want journalling on an usb stick or memory card.
But SSDs are also built on flash memory technology; do you disable
journalling on those as well?
No, they have wear levelling, and an expected lifetime with normal usage patterns that is quite long.
Exactly. SSDs algorithm and processing power (I read of an
SSD yesterday with a five core ARM processor in it), ensures
that the entire wear life of the device (number of cells times cycles)
is harvested. USB sticks don't even come remotely close to that. Some
USB sticks, don't even seem to follow what technical information
is available for them. Either their flash chips are entire crap
(should have been thrown out at flash factory), or, something
is very wrong with the controller.
I just realized I have an nvme with 72713 hours of use. Probably the first one I bought.
=== START OF INFORMATION SECTION ===
Model Family:     SandForce Driven SSDs
Device Model:     KINGSTON SMS200S3120G
Serial Number:    ...
LU WWN Device Id: 5 0026b7 26901494e
Firmware Version: 608ABBF0
User Capacity:    120,034,123,776 bytes [120 GB]
Sector Size:      512 bytes logical/physical
Rotation Rate:    Solid State Device
TRIM Command:     Available
Device is:        In smartctl database 7.3/5528
ATA Version is:   ATA8-ACS, ACS-2 T13/2015-D revision 3
SATA Version is:  SATA 3.0, 6.0 Gb/s (current: 6.0 Gb/s)
Local Time is:    Sat Mar 22 13:14:02 2025 CET
SMART support is: Available - device has SMART capability.
SMART support is: Enabled
...
SMART Attributes Data Structure revision number: 10
ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
  1 Raw_Read_Error_Rate     0x0032   095   095   050    Old_age   Always       -       0/38481593
  5 Retired_Block_Count     0x0033   100   100   003    Pre-fail  Always       -       0
  9 Power_On_Hours_and_Msec 0x0032   017   017   000    Old_age   Always       -       72713h+43m+19.000s
...
Post by Paul
Post by Carlos E.R.
SMART Self-test log structure revision number 1
Num  Test_Description    Status                  Remaining  LifeTime(hours)  LBA_of_first_error
# 1  Short offline       Completed without error       00%      7178         -
# 2  Extended offline    Completed without error       00%      7168         -
# 3  Short offline       Completed without error       00%      7166         -
That's amazing, that a 120GB drive is still alive. Some of those
die due to firmware issues.
Oh.
Post by Paul
it could be a SATA type NVME, rather than a PCIe.
This one has the small connector directly on the PCB. The first one I
saw. But the interesting thing is that it identifies as /dev/sda, not
/dev/nvme0n1
Post by Paul
The entry in /dev should help you identify what it is listed under.
Ah. Well, /dev/sda.
Post by Paul
As far as I know, Sandforce did compressing controllers for SATA,
and Kingston was their major customer. I could not tell you
whether Sandforce was still in business or not.
--
Cheers, Carlos.
Lawrence D'Oliveiro
2025-03-24 23:20:34 UTC
Reply
Permalink
Post by Carlos E.R.
Post by Lawrence D'Oliveiro
But SSDs are also built on flash memory technology; do you disable
journalling on those as well?
No, they have wear levelling, and an expected lifetime with normal usage
patterns that is quite long.
Fun fact: SSDs at a low level resemble what’s called a “log-structured”
filesystem. This is what happens when you have a filesystem that is all
journal, getting rid of the conventional filesystem part altogether.
Carlos E.R.
2025-03-25 11:55:08 UTC
Reply
Permalink
Post by Lawrence D'Oliveiro
Post by Carlos E.R.
Post by Lawrence D'Oliveiro
But SSDs are also built on flash memory technology; do you disable
journalling on those as well?
No, they have wear levelling, and an expected lifetime with normal usage
patterns that is quite long.
Fun fact: SSDs at a low level resemble what’s called a “log-structured”
filesystem. This is what happens when you have a filesystem that is all
journal, getting rid of the conventional filesystem part altogether.
Interesting.

That could be used to design a different filesystem, perhaps.
--
Cheers, Carlos.
Lawrence D'Oliveiro
2025-03-26 00:27:26 UTC
Reply
Permalink
Post by TJ
Post by Lawrence D'Oliveiro
Post by Carlos E.R.
Post by Lawrence D'Oliveiro
But SSDs are also built on flash memory technology; do you disable
journalling on those as well?
No, they have wear levelling, and an expected lifetime with normal
usage patterns that is quite long.
Fun fact: SSDs at a low level resemble what’s called a “log-structured”
filesystem. This is what happens when you have a filesystem that is all
journal, getting rid of the conventional filesystem part altogether.
Interesting.
That could be used to design a different filesystem, perhaps.
Note the point, though: the journal itself provides the wear-levelling.
Anssi Saari
2025-03-20 09:07:37 UTC
Reply
Permalink
Post by Ant
Post by Anssi Saari
Post by Carlos E.R.
exFAT is better, but few TV sets support it, while many support NTFS.
So how is exFAT better then? As a portable FS?
exFAT can handle bigger files and partitions. FAT32 can't. See
https://duckduckgo.com/?q=exfat+vs.+fat32 for the details. Also, exFAT
is better for portabilities.
Wow, way to go off on a weird tangent. I specifically wanted Carlos to
explain why he thinks exFAT is better than NTFS for portability since he
seemed to contradict himself, saying, as he did, that "few TV sets
support it".

Carlos' answer was, apparently, "yes".
Carlos E.R.
2025-03-20 11:29:24 UTC
Reply
Permalink
Post by Anssi Saari
Post by Ant
Post by Anssi Saari
Post by Carlos E.R.
exFAT is better, but few TV sets support it, while many support NTFS.
So how is exFAT better then? As a portable FS?
exFAT can handle bigger files and partitions. FAT32 can't. See
https://duckduckgo.com/?q=exfat+vs.+fat32 for the details. Also, exFAT
is better for portabilities.
Wow, way to go off on a weird tangent. I specifically wanted Carlos to
explain why he thinks exFAT is better than NTFS for portability since he
seemed to contradict himself, saying, as he did, that "few TV sets
support it".
Carlos' answer was, apparently, "yes".
exFAT is better that NTFS on removable flash media such as sticks or
memory cards, because it was designed for that usage. It is better than
NTFS or FAT for exchanging files between Windows and Linux (because it
is really supported on Linux, and supports large files).

However, being better is pointless if the destination machine does not
support it.
--
Cheers, Carlos.
Paul
2025-03-19 18:14:28 UTC
Reply
Permalink
Post by Anssi Saari
Post by Carlos E.R.
exFAT is better, but few TV sets support it, while many support NTFS.
So how is exFAT better then? As a portable FS?
It is FAT32 with large clusters and larger allowed number of files.

https://en.wikipedia.org/wiki/ExFAT

It may have been added to WinXP as an IFS (Installable File System,
like a FUSE but for Windows).

Paul
Carlos E.R.
2025-03-19 20:18:41 UTC
Reply
Permalink
Post by Anssi Saari
Post by Carlos E.R.
exFAT is better, but few TV sets support it, while many support NTFS.
So how is exFAT better then? As a portable FS?
Yes, if your destination machine supports it.
--
Cheers, Carlos.
Anssi Saari
2025-03-20 08:51:43 UTC
Reply
Permalink
Post by Carlos E.R.
Post by Anssi Saari
Post by Carlos E.R.
exFAT is better, but few TV sets support it, while many support NTFS.
So how is exFAT better then? As a portable FS?
Yes, if your destination machine supports it.
Yes? I asked how and you answer yes?
Carlos E.R.
2025-03-20 11:22:55 UTC
Reply
Permalink
Post by Anssi Saari
Post by Carlos E.R.
Post by Anssi Saari
Post by Carlos E.R.
exFAT is better, but few TV sets support it, while many support NTFS.
So how is exFAT better then? As a portable FS?
Yes, if your destination machine supports it.
Yes? I asked how and you answer yes?
I don't understand what "how" means in this context.
--
Cheers, Carlos.
TJ
2025-03-21 00:35:19 UTC
Reply
Permalink
Post by Carlos E.R.
Post by Anssi Saari
Post by Carlos E.R.
exFAT is better, but few TV sets support it, while many support NTFS.
So how is exFAT better then? As a portable FS?
Yes, if your destination machine supports it.
Ay, there's the rub.

I have a ATSC->NTSC TV converter that has a usb port and a PVR function.
It can also play various formats of video files from a device in that
port. It can work with FAT32 or NTFS. That's it. No exceptions.

ExFAT would probably work better, if only it supported it.

TJ
Carlos E.R.
2025-03-21 10:55:34 UTC
Reply
Permalink
Post by TJ
Post by Carlos E.R.
Post by Anssi Saari
Post by Carlos E.R.
exFAT is better, but few TV sets support it, while many support NTFS.
So how is exFAT better then? As a portable FS?
Yes, if your destination machine supports it.
Ay, there's the rub.
I have a ATSC->NTSC TV converter that has a usb port and a PVR function.
It can also play various formats of video files from a device in that
port. It can work with FAT32 or NTFS. That's it. No exceptions.
ExFAT would probably work better, if only it supported it.
Exactly my problem.

On the other hand, I find that TV sets support for playing media is
terrible. Some cases they do not support subtitles aka captions, others
the forward/backwards keys are so slow as being unusable. So I end by
having an old laptop permanently attached to the TV.
--
Cheers, Carlos.
Joerg Walther
2025-03-21 15:24:11 UTC
Reply
Permalink
Post by Carlos E.R.
On the other hand, I find that TV sets support for playing media is
terrible. Some cases they do not support subtitles aka captions, others
the forward/backwards keys are so slow as being unusable. So I end by
having an old laptop permanently attached to the TV.
I use a Raspberry Pi with Libreelec for this, this probably is the best
media center around, it can even download subtitles while watching and
it plays every file I throw at it. An old Windows Media Center remote
worked out of the box, which came very handy.

-jw-
--
And now for something completely different...
Lawrence D'Oliveiro
2025-03-22 06:58:07 UTC
Reply
Permalink
An old Windows Media Center remote worked out of the box, which came
very handy.
What happened to Windows Media Center?

Killed off by Linux.
Paul
2025-03-22 08:29:28 UTC
Reply
Permalink
Post by Lawrence D'Oliveiro
An old Windows Media Center remote worked out of the box, which came
very handy.
What happened to Windows Media Center?
Killed off by Linux.
Uh, not really.

when it comes to "continuity" on high tech fluff, the
bean counters hate "third party expense". For example,
even if an MPEG2 license from MPEG-LA costs a dollar
a node, that's a "whoa! hold on there" issue for the
bean counters at Microsoft. They want a counter-balancing
income if you do that.

Right now on Windows, it needs a HEVC license from the
Microsoft Store, so that HEIC can be decoded. Rather than
Microsoft pay that out of their own pocket, you buy the
item from the Microsoft Store, and that gives you the
CODEC needed. If they didn't do that, they'd be sued.

Linux doesn't get sued, because they are giving away
the software and not making money from it. the Linux
overhead expense is different.

As "compensation" for Media Center, for a short time,
on the next OS where Media Center was discontinued,
you were given two MPEG2 CODECS "for free". To Microsoft
then, that represented the "value" to them, of the
removed software.

Another expense for Media Center, was the Guide Data feed
per user. Linux doesn't have Guide Data. Professional
Guide Data always costs money. You can license Guide Data
from a TV Network, for around $50K per annum. Then,
you chop that up, and bill individual customers, to make
your $50K back. When Guide Data sources go out of business,
it's because they could not sell enough units at $25 per year.
Microsoft was paying someone else for the Guide Data, while
Media Center was available. I used to have Guide Data downloads
every day on the Test Machine (which was running Media Center
for a while as a demo, so I could note the missing bits
in USENET posts). For example, in Canada, the digital TV
side of Media Center, did not work, unless you got some
files from a private citizen in Canada, who had figured out
how to fix it. That's how I got mine running.

Media Center sank under its own weight. Linux had nothing
to do with the business decisions (overhead costs). If you
buy the WinTV software from Hauppauge, that's another way
to record TV programs. I don't know if the Guide Data for
that still works or not.

Paul
Lawrence D'Oliveiro
2025-03-22 22:00:13 UTC
Reply
Permalink
Post by Paul
Post by Lawrence D'Oliveiro
What happened to Windows Media Center?
Killed off by Linux.
Uh, not really.
Right now on Windows, it needs a HEVC license from the Microsoft Store,
so that HEIC can be decoded. Rather than Microsoft pay that out of their
own pocket, you buy the item from the Microsoft Store, and that gives
you the CODEC needed. If they didn't do that, they'd be sued.
Linux doesn't get sued, because they are giving away the software and
not making money from it. the Linux overhead expense is different.
As "compensation" for Media Center, for a short time,
on the next OS where Media Center was discontinued,
you were given two MPEG2 CODECS "for free". To Microsoft then, that
represented the "value" to them, of the removed software.
Another expense for Media Center, was the Guide Data feed per user.
Linux doesn't have Guide Data. Professional Guide Data always costs
money.
Funny, you said Windows Media Center wasn’t killed by Linux, and then go
on to list a long catalogue of reasons why it was.
Paul
2025-03-23 06:01:29 UTC
Reply
Permalink
Post by Lawrence D'Oliveiro
Funny, you said Windows Media Center wasn’t killed by Linux, and then go
on to list a long catalogue of reasons why it was.
It wasn't killed by linus, it was killed by overhead cost.

If it was costing them nothing (no bills coming in for MPEG-LA
licenses or no bill for Guide Data), then it would still be
available today.

A similar thing happened to the NVidia chipset that had five DSP
cores in the Southbridge. One of the cores did AC3 encoding with
low latency. A great little toy, for some people with SPDIF connected
AV receivers. Well, that did not get included in the very next chipset,
and the reason, was the license cost per chip (even if the customer
wasn't using the feature, the AC3 license had to be paid). So the
DSP feature wasn't used any more.

Any time there is a "line-item expense" associated with an activity,
a bean counter will stop it. That's how business works. As the
product manager, you have to show the bean counter, how you are
making the money back in some way. And one of the schemes for that,
is to make the customer buy the license when they want one
(from a thing such as an online Store).

On the first generation of a product idea, you can write
"Promotional cost" next to the line item, for its year of
introduction. But after a product is mature, you have to
justify every line-item expense. And that's how your
product gets canceled.

Paul
Lawrence D'Oliveiro
2025-03-24 00:43:28 UTC
Reply
Permalink
Post by Lawrence D'Oliveiro
Funny, you said Windows Media Center wasn’t killed by Linux, and then
go on to list a long catalogue of reasons why it was.
It wasn't killed by linu[x], it was killed by overhead cost.
Another area in which Linux had the advantage, and was able to take over
the market thereby.

You just keep coming up with more and more reasons why my point stands.
TJ
2025-03-21 20:37:10 UTC
Reply
Permalink
Post by Carlos E.R.
Post by TJ
Post by Carlos E.R.
Post by Anssi Saari
Post by Carlos E.R.
exFAT is better, but few TV sets support it, while many support NTFS.
So how is exFAT better then? As a portable FS?
Yes, if your destination machine supports it.
Ay, there's the rub.
I have a ATSC->NTSC TV converter that has a usb port and a PVR
function. It can also play various formats of video files from a
device in that port. It can work with FAT32 or NTFS. That's it. No
exceptions.
ExFAT would probably work better, if only it supported it.
Exactly my problem.
On the other hand, I find that TV sets support for playing media is
terrible. Some cases they do not support subtitles aka captions, others
the forward/backwards keys are so slow as being unusable. So I end by
having an old laptop permanently attached to the TV.
I have My converter box in my bedroom, playing through an old Commodore
1701 monitor. Better resolution than an analog TV set of that era, but
not as good as HD.

Used to belong to an Atari User Group for use with their 8-bit
computers. There was a time, back in the day, when I was an Atari XL/XE
power user. My XE had been upgraded to have a whole 320KB of RAM.

It was a simpler time...

I miss subtitles sometimes, but not very much. For my purposes, as long
as the video is as watchable as what I had with my old NTSC TV, I'm happy.

TJ
Lawrence D'Oliveiro
2025-03-22 06:57:29 UTC
Reply
Permalink
Post by Carlos E.R.
On the other hand, I find that TV sets support for playing media is
terrible.
Get something like a Kodi box.
Carlos E.R.
2025-03-22 13:00:15 UTC
Reply
Permalink
Post by Lawrence D'Oliveiro
Post by Carlos E.R.
On the other hand, I find that TV sets support for playing media is
terrible.
Get something like a Kodi box.
Well, the laptop I already have, and would be gathering dust otherwise.
Actually, sometimes I run kodi on it.

It is funny, some videos barely run on the laptop using VLC; however,
they run fine using kodi.

Even if I recode those videos using ffmpeg, VLC can't play them. Sound
is good, video stalls.
--
Cheers, Carlos.
TJ
2025-03-22 13:42:26 UTC
Reply
Permalink
Post by Carlos E.R.
Post by Lawrence D'Oliveiro
Post by Carlos E.R.
On the other hand, I find that TV sets support for playing media is
terrible.
Get something like a Kodi box.
Well, the laptop I already have, and would be gathering dust otherwise.
Actually, sometimes I run kodi on it.
It is funny, some videos barely run on the laptop using VLC; however,
they run fine using kodi.
Even if I recode those videos using ffmpeg, VLC can't play them. Sound
is good, video stalls.
Have you tried recoding with Handbrake? It has worked for me in the past.

TJ
Carlos E.R.
2025-03-22 14:20:28 UTC
Reply
Permalink
Post by TJ
Post by Carlos E.R.
Post by Lawrence D'Oliveiro
Post by Carlos E.R.
On the other hand, I find that TV sets support for playing media is
terrible.
Get something like a Kodi box.
Well, the laptop I already have, and would be gathering dust
otherwise. Actually, sometimes I run kodi on it.
It is funny, some videos barely run on the laptop using VLC; however,
they run fine using kodi.
Even if I recode those videos using ffmpeg, VLC can't play them. Sound
is good, video stalls.
Have you tried recoding with Handbrake? It has worked for me in the past.
I haven't, but most tools use the same codec libraries.

I tried these:

time ffmpeg -i Source\ x265\ .mkv \
-map 0 -vf scale=1920:-1 -c:v libx264 -preset slow -crf 22 -tune fastdecode -profile:v baseline -c:a copy -c:s copy \
Dest\ baseline.mkv

time ffmpeg -i Source\ x265\ .mkv \
-map 0 -vf scale=1920:-1 -c:v libx264 -preset slow -crf 22 -tune fastdecode -profile:v main -c:a copy -c:s copy \
Dest\ -\ main.mkv

time ffmpeg -i Source\ x265\ .mkv \
-map 0 -vf scale=1920:-1 -c:v mpeg4 -vtag xvid -qscale:v 3 -c:a copy -c:s copy \
Dest\ xvid.mkv


I have trouble when the video uses x265. x264 is fine, so I was trying to encode to x264. VLC on that laptop has trouble with those (I do the recoding in another machine that is powerful).


But if kodi can display the video, it is less effort and resources to watch the movies in kodi. It is just curious, as both vlc and kodi link the same libx265.so libraries.
--
Cheers, Carlos.
Paul
2025-03-22 18:00:45 UTC
Reply
Permalink
Post by Carlos E.R.
Post by TJ
Post by Lawrence D'Oliveiro
Post by Carlos E.R.
On the other hand, I find that TV sets support for playing media is
terrible.
Get something like a Kodi box.
Well, the laptop I already have, and would be gathering dust otherwise. Actually, sometimes I run kodi on it.
It is funny, some videos barely run on the laptop using VLC; however, they run fine using kodi.
Even if I recode those videos using ffmpeg, VLC can't play them. Sound is good, video stalls.
Have you tried recoding with Handbrake? It has worked for me in the past.
I haven't, but most tools use the same codec libraries.
time ffmpeg -i Source\ x265\ .mkv \
    -map 0 -vf scale=1920:-1  -c:v libx264 -preset slow -crf 22 -tune fastdecode -profile:v baseline  -c:a copy -c:s copy \
    Dest\ baseline.mkv
time ffmpeg -i Source\ x265\ .mkv \
    -map 0 -vf scale=1920:-1  -c:v libx264 -preset slow -crf 22 -tune fastdecode -profile:v main  -c:a copy -c:s copy \
    Dest\ -\ main.mkv
time ffmpeg -i Source\ x265\ .mkv \
    -map 0 -vf scale=1920:-1  -c:v mpeg4 -vtag xvid -qscale:v 3  -c:a copy -c:s copy \
    Dest\ xvid.mkv
I have trouble when the video uses x265. x264 is fine, so I was trying to encode to x264. VLC on that laptop has trouble with those (I do the recoding in another machine that is powerful).
But if kodi can display the video, it is less effort and resources to watch the movies in kodi. It is just curious, as both vlc and kodi link the same libx265.so libraries.
Take it all the way back to RAW, then re-code it. Then you can
do bidirectional encoding if you want (for better random seek
behavior). Or for that matter, re-code with only keyframes,
high bitrate, and (almost no) compression :-)

One way to go back to a RAW format, is to store the video frames
as individual pictures in a folder. I've done that before, as
part of experimenting with video.

But don't expect the sound track to stay synchronized. Sound
only remains synced, if the video track and the audio track
have the original timestamps. What you will find on practical
videos, is the sound track speeds up or slows down at random.
If you use a video editor and attempt to "slide" the
sound track with respect to the video track, yes, you can
align the sound at a selected point on the video, but other
parts of the video are then improperly aligned.

*******

You can tell the FFMPEG library to use the hardware decoder
in the iGPU, like the Intel QuickSync encoder/decoder or
the like. You don't always have to use the software decoder
for this.

When you re-code using FFMPEG, you can spec hardware decoding
on input of the video, then software re-encode with the rest
of the FFMPEG command. You can even do the entire job
in hardware (about 10x speedup over software method).
But since NVENC and NVDEC are not switched on in the
Linux FFMPEG, you can recompile from source and
use ./configure to add back NVENC and NVDEC.

Doing this stuff, is more of a "hobby" than a ten minute
project :-) All I wanted to learn, is how long would it
take me to fix a video. My answer to that is
"book two weeks of your time for it".

Paul
Carlos E.R.
2025-03-22 19:34:21 UTC
Reply
Permalink
Post by Paul
Post by Carlos E.R.
Post by TJ
Post by Lawrence D'Oliveiro
Post by Carlos E.R.
On the other hand, I find that TV sets support for playing media is
terrible.
Get something like a Kodi box.
Well, the laptop I already have, and would be gathering dust otherwise. Actually, sometimes I run kodi on it.
It is funny, some videos barely run on the laptop using VLC; however, they run fine using kodi.
Even if I recode those videos using ffmpeg, VLC can't play them. Sound is good, video stalls.
Have you tried recoding with Handbrake? It has worked for me in the past.
I haven't, but most tools use the same codec libraries.
time ffmpeg -i Source\ x265\ .mkv \
    -map 0 -vf scale=1920:-1  -c:v libx264 -preset slow -crf 22 -tune fastdecode -profile:v baseline  -c:a copy -c:s copy \
    Dest\ baseline.mkv
time ffmpeg -i Source\ x265\ .mkv \
    -map 0 -vf scale=1920:-1  -c:v libx264 -preset slow -crf 22 -tune fastdecode -profile:v main  -c:a copy -c:s copy \
    Dest\ -\ main.mkv
time ffmpeg -i Source\ x265\ .mkv \
    -map 0 -vf scale=1920:-1  -c:v mpeg4 -vtag xvid -qscale:v 3  -c:a copy -c:s copy \
    Dest\ xvid.mkv
I have trouble when the video uses x265. x264 is fine, so I was trying to encode to x264. VLC on that laptop has trouble with those (I do the recoding in another machine that is powerful).
But if kodi can display the video, it is less effort and resources to watch the movies in kodi. It is just curious, as both vlc and kodi link the same libx265.so libraries.
Take it all the way back to RAW, then re-code it. Then you can
do bidirectional encoding if you want (for better random seek
behavior). Or for that matter, re-code with only keyframes,
high bitrate, and (almost no) compression :-)
One way to go back to a RAW format, is to store the video frames
as individual pictures in a folder. I've done that before, as
part of experimenting with video.
But don't expect the sound track to stay synchronized. Sound
only remains synced, if the video track and the audio track
have the original timestamps. What you will find on practical
videos, is the sound track speeds up or slows down at random.
If you use a video editor and attempt to "slide" the
sound track with respect to the video track, yes, you can
align the sound at a selected point on the video, but other
parts of the video are then improperly aligned.
Too much work... If kodi can play them, I'm happy.

Some movies have been encoded for 4K and are huge, though.
Post by Paul
*******
You can tell the FFMPEG library to use the hardware decoder
in the iGPU, like the Intel QuickSync encoder/decoder or
the like. You don't always have to use the software decoder
for this.
When you re-code using FFMPEG, you can spec hardware decoding
on input of the video, then software re-encode with the rest
of the FFMPEG command. You can even do the entire job
in hardware (about 10x speedup over software method).
But since NVENC and NVDEC are not switched on in the
Linux FFMPEG, you can recompile from source and
use ./configure to add back NVENC and NVDEC.
The CPU in the desktop machine is AMD, and so is the GPU as well.

If you are curious:

Telcontar:~ # inxi -C -GSaz --za --vs
inxi 3.3.31-00 (2023-11-02)
System:
Kernel: 6.4.0-150600.23.42-default arch: x86_64 bits: 64 compiler: gcc
v: 7.5.0 clocksource: tsc available: hpet,acpi_pm
parameters: BOOT_IMAGE=/vmlinuz-6.4.0-150600.23.42-default
root=UUID=<filter> resume=/dev/disk/by-label/nvme-swap splash=verbose
verbose
Desktop: Xfce v: 4.20.0 tk: Gtk v: 3.24.43 info: xfce4-panel wm: xfwm
v: 4.20.0 dm: 1: GDM v: 45.0.1 2: SDDM note: stopped
Distro: openSUSE Leap 15.6
CPU:
Info: model: AMD Ryzen 5 3600X socket: AM4 bits: 64 type: MT MCP arch: Zen 2
gen: 3 level: v3 note: check built: 2020-22 process: TSMC n7 (7nm)
family: 0x17 (23) model-id: 0x71 (113) stepping: 0 microcode: 0x8701034
Topology: cpus: 1x cores: 6 tpc: 2 threads: 12 smt: enabled cache:
L1: 384 KiB desc: d-6x32 KiB; i-6x32 KiB L2: 3 MiB desc: 6x512 KiB
L3: 32 MiB desc: 2x16 MiB
Speed (MHz): avg: 2350 high: 2800 min/max: 2200/4409 boost: enabled
base/boost: 3800/4400 scaling: driver: acpi-cpufreq governor: ondemand
volts: 1.1 V ext-clock: 100 MHz cores: 1: 2800 2: 2800 3: 2200 4: 2800
5: 2200 6: 2200 7: 2200 8: 2200 9: 2200 10: 2200 11: 2200 12: 2200
bogomips: 91203
Flags: avx avx2 ht lm nx pae sse sse2 sse3 sse4_1 sse4_2 sse4a ssse3 svm
Vulnerabilities: <filter>
Graphics:
Device-1: AMD Ellesmere [Radeon RX 470/480/570/570X/580/580X/590]
vendor: Micro-Star MSI driver: amdgpu v: kernel arch: GCN-4
code: Arctic Islands process: GF 14nm built: 2016-20 pcie: gen: 3
speed: 8 GT/s lanes: 16 ports: active: DVI-D-1 empty: DP-1, DP-2,
HDMI-A-1, HDMI-A-2 bus-ID: 27:00.0 chip-ID: 1002:67df class-ID: 0300
temp: 37.0 C
Display: unspecified server: X.Org v: 1.21.1.11 with: Xwayland v: 24.1.1
compositor: xfwm v: 4.20.0 driver: X: loaded: amdgpu
unloaded: fbdev,modesetting,vesa dri: radeonsi gpu: amdgpu
display-ID: :0.0 screens: 1
Screen-1: 0 s-res: 1920x1080 s-dpi: 96 s-size: 509x286mm (20.04x11.26")
s-diag: 584mm (22.99")
Monitor-1: DVI-D-1 mapped: DVI-D-0 model: Acer H243HX serial: <filter>
built: 2009 res: 1920x1080 hz: 60 dpi: 92 gamma: 1.2
size: 531x298mm (20.91x11.73") diag: 604mm (23.8") ratio: 16:9 modes:
max: 1920x1080 min: 720x400
API: OpenGL v: 4.6 vendor: amd mesa v: 23.3.4 glx-v: 1.4 es-v: 3.2
direct-render: yes renderer: AMD Radeon RX 580 Series (radeonsi polaris10
LLVM 17.0.6 DRM 3.57 6.4.0-150600.23.42-default) device-ID: 1002:67df
memory: 7.81 GiB unified: no
API: Vulkan v: 1.3.275 layers: 1 device: 0 type: discrete-gpu name: AMD
Radeon RX 580 Series (RADV POLARIS10) driver: N/A device-ID: 1002:67df
surfaces: xcb,xlib
API: EGL Message: EGL data requires eglinfo. Check --recommends.
Telcontar:~ #
Post by Paul
Doing this stuff, is more of a "hobby" than a ten minute
project :-) All I wanted to learn, is how long would it
take me to fix a video. My answer to that is
"book two weeks of your time for it".
Argh!
Post by Paul
Paul
--
Cheers, Carlos.
Lawrence D'Oliveiro
2025-03-22 22:02:30 UTC
Reply
Permalink
Post by TJ
Post by Carlos E.R.
Even if I recode those videos using ffmpeg, VLC can't play them. Sound
is good, video stalls.
Have you tried recoding with Handbrake? It has worked for me in the past.
HandBrake is built on a forked subset of the FFmpeg libraries.

FFmpeg itself offers a host of options for encoding; I’m sure you’ll find
something that works.

Before I got my Kodi box, I was running a proprietary WD media streamer
which was very finicky about things. For example, it would not play MPEG-4
H264 encoded with a pixel format of 4:4:4 (the FFmpeg default), but it was
fine with 4:2:0.
Carlos E.R.
2025-03-22 22:22:25 UTC
Reply
Permalink
Post by Lawrence D'Oliveiro
Post by TJ
Post by Carlos E.R.
Even if I recode those videos using ffmpeg, VLC can't play them. Sound
is good, video stalls.
Have you tried recoding with Handbrake? It has worked for me in the past.
HandBrake is built on a forked subset of the FFmpeg libraries.
FFmpeg itself offers a host of options for encoding; I’m sure you’ll find
something that works.
Before I got my Kodi box, I was running a proprietary WD media streamer
which was very finicky about things. For example, it would not play MPEG-4
H264 encoded with a pixel format of 4:4:4 (the FFmpeg default), but it was
fine with 4:2:0.
Hum. I have never played with the pixel format.
--
Cheers, Carlos.
Lawrence D'Oliveiro
2025-03-24 00:42:13 UTC
Reply
Permalink
Post by Carlos E.R.
Post by Lawrence D'Oliveiro
Before I got my Kodi box, I was running a proprietary WD media
streamer which was very finicky about things. For example, it would
not play MPEG-4 H264 encoded with a pixel format of 4:4:4 (the
FFmpeg default), but it was fine with 4:2:0.
Hum. I have never played with the pixel format.
There are many, many options to play with in video encoding in FFmpeg.
They exist not (just) for fun, but because they can make a real difference
in quality as well as compatibility.
Lawrence D'Oliveiro
2025-03-17 23:18:16 UTC
Reply
Permalink
Post by Anssi Saari
I've actually found NTFS on a USB SSD to be surprisingly widely
supported on media players and TVs and such. I've used it on Android
too.
None of which are running Windows. They would all be running some variety
of Linux kernel (Android definitely so).
R.Wieser
2025-03-15 08:08:51 UTC
Reply
Permalink
Lawrence,
Windows doesn't have the idea that any storage device could be
hot-pluggable, does it.
Not *any* device, no. And that has to do with caching as well as
electronics.

Connecting signal wires before the ground terminal will often cause the
logics attached to it to "see" stuff that isn't there. Same kind of problem
when the power terminal connects and the device is, electrically,
initializing.

Regards,
Rudy Wieser
Paul
2025-03-15 17:37:48 UTC
Reply
Permalink
Post by R.Wieser
Lawrence,
Windows doesn't have the idea that any storage device could be
hot-pluggable, does it.
Not *any* device, no. And that has to do with caching as well as
electronics.
Connecting signal wires before the ground terminal will often cause the
logics attached to it to "see" stuff that isn't there. Same kind of problem
when the power terminal connects and the device is, electrically,
initializing.
Regards,
Rudy Wieser
And that's why, if you look at the design of hot-plug devices,
you will notice some contacts are longer than others and this
is intended to make them "touch first" on insertion.

As an illustration, the SATA data 7 pin section on the hard drive
looks like this. The grounds are longer than the diff pairs for
transmit and receive (TX+,TX- and RX+,RX-). When you insert the
SATA data connector, the grounds touch first.

G G G
G D D G D D G

On electronics, you can have multiple levels of contact heights,
intended to "sequence" the electronics when plugged in.

Paul
R.Wieser
2025-03-15 18:22:06 UTC
Reply
Permalink
Paul,
Post by Paul
And that's why, if you look at the design of hot-plug
devices, you will notice some contacts are longer than
others and this is intended to make them "touch first"
on insertion.
I know, you know, and most likely others do too.

And its the reason why you don't try to hot-plug a PATA drive. :-)

Regards,
Rudy Wieser
Paul
2025-03-15 18:57:43 UTC
Reply
Permalink
Post by R.Wieser
Paul,
Post by Paul
And that's why, if you look at the design of hot-plug
devices, you will notice some contacts are longer than
others and this is intended to make them "touch first"
on insertion.
I know, you know, and most likely others do too.
And its the reason why you don't try to hot-plug a PATA drive. :-)
Regards,
Rudy Wieser
You *can* hot plug a ribbon cable drive.

You need the right kind of housing with the
right kind of connectors on either side, to do it.
It can only be done safely, by using a housed drive,
that slides into a special tray. The special tray
goes to the end-connector on the IDE ribbon (jumpered Master)
and the special tray stays attached to that cable
all the time. The IDE housed drive, slides in or out
of the tray. There is a little ceremony, during removal.

This was not a popular activity at the time,
and you may have a lot of trouble finding
a Google image of the items in question.

This article kinda hints at the detail.

https://en.wikipedia.org/wiki/Parallel_ATA

"Although CF can be hot-pluggable with additional design methods,
by default when wired directly to an ATA interface, it is
not intended to be hot-pluggable."

It's the same for the PATA drive. The ribbon cable, considered by itself,
is as the spec summary at the top says:

General specifications

Hot pluggable No

But, by means of redesigning the connectors that mate between the
housing that slides in and out, and the tray it rests in, it
can be made hot-plug compatible.

I had *absolutely no interest* in testing this, and did not
buy samples of the housing as a result :-) It's like a bar bet,
doing stuff like this. But a commercial interest, did sell
housings to do this very thing. Even at the time these were for
sale, the documentation was very poor. You could hardly
tell what they were doing, at the interface level.

Paul
R.Wieser
2025-03-15 20:06:02 UTC
Reply
Permalink
Paul,
Post by Paul
You *can* hot plug a ribbon cable drive.
You need the right kind of housing with the
right kind of connectors on either side,
:-) You know that, I know that.
Post by Paul
But, by means of redesigning the connectors that mate between
the housing that slides in and out, and the tray it rests in, it
can be made hot-plug compatible.
Making sure ground comes first is one part of the equation. Having
electronics which could withstand the (minimal?) surges that came with
inserting/extracting while powered is another.
Post by Paul
I had *absolutely no interest* in testing this, and did
not buy samples of the housing as a result :-)
For a while (long ago) I had several operating systems on seperate drives in
trays. Though in this case (changing the OS) I always powered-down first.

Regards,
Rudy Wieser
Paul
2025-03-11 23:56:13 UTC
Reply
Permalink
Post by David W. Hodgins
Post by Daniel70
Some time ago, I backed up my Laptop's 500GB internal HD to an 2TB
external HD using (I forget) ... It might have been 'dd' but I doubt it.
Depending on how you did it, the backup may be in a file in a partition on the 2TB drive,
or the partition table may have been copied so that the partitions on the 2TB drive are
the same as they were on the 500GB drive.
Post by Daniel70
However, the 500GB Internals Image took up 500GB on the External (i.e. a
byte-by-byte image, even the empty bytes, apparently!).
Now, when I look at that External HD using my Win 11 Desk-top, it
doesn't 'see' anything.
WIndows should see an "unknown" file system in the existing partition(s).
Post by Daniel70
Is this because Linux is at a 'Higher'/'deeper' level than Windows can
'see'??
M$ doesn't want to make it easy for windows users to use anything else.
Post by Daniel70
If I back-up this Win-11 Desktop to the same External HD, is there a
possibility that Win-11 will write itself over the Linux Image?? Or is
Win-11 able to detect that there is 'something' there so will go looking
for the next available UNUSED portion of the External HD??
It really matters how the backup was done. Windows may overwrite it or it may
allow you to create new partitions on the drive. I don't trust windows for anything,
and haven't used it much since XP. I occasionally troubleshoot things for others,
but try to avoid it as much as I can.
Don't write anything to the drive until you know exactly what is on there. Working
with low level tools like dd make it easy to wipe out data with a single typo.
Regards, Dave Hodgins
A Macrium Rescue CD (a windows ecosystem product), can back up
Windows file systems, as well as EXTm file systems. Depending
on which version you use, it may not be compatible with C12 feature
(because C12 feature came after the Macrium was released). It might
take a paid version of Macrium to handle that (which is why people
create "features" this late in the life of a filesystem).

*Nothing* prevents a Windows ecosystem backup product from
backing up (at-rest) ZFS, BTRFS, EXT, ReiserFS and so on.
All that is needed, is enough of a driver to measure and determine
which inodes need to be recorded (and restored later), in the
backup .img file .

I finally got a hit, on a list that gives some notion of capability.
For example, Disk Genius (untested, country of origin unknown). There
is a lot of URL monkey business in the industry, call it astroturfing
and you never know when a new company appears, whether it isn't just
one of the older companies re-branding for profit.

https://en.wikipedia.org/wiki/Comparison_of_disk_cloning_software

Paul
Richard Kettlewell
2025-03-11 18:16:47 UTC
Reply
Permalink
Post by Daniel70
Some time ago, I backed up my Laptop's 500GB internal HD to an 2TB
external HD using (I forget) ... It might have been 'dd' but I doubt it.
A backup that you don’t know how to restore isn’t really any use.
Post by Daniel70
However, the 500GB Internals Image took up 500GB on the External
(i.e. a byte-by-byte image, even the empty bytes, apparently!).
Now, when I look at that External HD using my Win 11 Desk-top, it
doesn't 'see' anything.
Look at it with a Linux box instead.
Post by Daniel70
If I back-up this Win-11 Desktop to the same External HD, is there a
possibility that Win-11 will write itself over the Linux Image?? Or is
Win-11 able to detect that there is 'something' there so will go
looking for the next available UNUSED portion of the External HD??
Not possible to say, since you don’t say how you made the backup.
--
https://www.greenend.org.uk/rjk/
Paul
2025-03-12 00:03:17 UTC
Reply
Permalink
Post by Richard Kettlewell
Post by Daniel70
Some time ago, I backed up my Laptop's 500GB internal HD to an 2TB
external HD using (I forget) ... It might have been 'dd' but I doubt it.
A backup that you don’t know how to restore isn’t really any use.
Post by Daniel70
However, the 500GB Internals Image took up 500GB on the External
(i.e. a byte-by-byte image, even the empty bytes, apparently!).
Now, when I look at that External HD using my Win 11 Desk-top, it
doesn't 'see' anything.
Look at it with a Linux box instead.
Post by Daniel70
If I back-up this Win-11 Desktop to the same External HD, is there a
possibility that Win-11 will write itself over the Linux Image?? Or is
Win-11 able to detect that there is 'something' there so will go
looking for the next available UNUSED portion of the External HD??
Not possible to say, since you don’t say how you made the backup.
sudo apt install disktype # An application having a small source size, but quite capable

sudo disktype /dev/sda # check a hard drive (needs elevation with sudo)

file my.img # Check if the format is recognizable
disktype my.img # If it is a dd img, display the partition information

Paul
Carlos E.R.
2025-03-12 14:12:04 UTC
Reply
Permalink
Post by Paul
Post by Richard Kettlewell
Post by Daniel70
Some time ago, I backed up my Laptop's 500GB internal HD to an 2TB
external HD using (I forget) ... It might have been 'dd' but I doubt it.
A backup that you don’t know how to restore isn’t really any use.
Post by Daniel70
However, the 500GB Internals Image took up 500GB on the External
(i.e. a byte-by-byte image, even the empty bytes, apparently!).
Now, when I look at that External HD using my Win 11 Desk-top, it
doesn't 'see' anything.
Look at it with a Linux box instead.
Post by Daniel70
If I back-up this Win-11 Desktop to the same External HD, is there a
possibility that Win-11 will write itself over the Linux Image?? Or is
Win-11 able to detect that there is 'something' there so will go
looking for the next available UNUSED portion of the External HD??
Not possible to say, since you don’t say how you made the backup.
sudo apt install disktype # An application having a small source size, but quite capable
Do you know where to find its source?

My distro doesn't have it.
--
Cheers, Carlos.
Paul
2025-03-12 18:56:08 UTC
Reply
Permalink
Post by Carlos E.R.
Post by Richard Kettlewell
Post by Daniel70
Some time ago, I backed up my Laptop's 500GB internal HD to an 2TB
external HD using (I forget) ... It might have been 'dd' but I doubt it.
A backup that you don’t know how to restore isn’t really any use.
Post by Daniel70
However, the 500GB Internals Image took up 500GB on the External
(i.e. a byte-by-byte image, even the empty bytes, apparently!).
Now, when I look at that External HD using my Win 11 Desk-top, it
doesn't 'see' anything.
Look at it with a Linux box instead.
Post by Daniel70
If I back-up this Win-11 Desktop to the same External HD, is there a
possibility that Win-11 will write itself over the Linux Image?? Or is
Win-11 able to detect that there is 'something' there so will go
looking for the next available UNUSED portion of the External HD??
Not possible to say, since you don’t say how you made the backup.
sudo apt install disktype      # An application having a small source size, but quite capable
Do you know where to find its source?
My distro doesn't have it.
https://disktype.sourceforge.net/

( https://sourceforge.net/projects/disktype/ )

Paul
bad sector
2025-03-17 11:49:47 UTC
Reply
Permalink
Post by Daniel70
Some time ago, I backed up my Laptop's 500GB internal HD to an 2TB
external HD using (I forget) ... It might have been 'dd' but I doubt it.
However, the 500GB Internals Image took up 500GB on the External (i.e. a
byte-by-byte image, even the empty bytes, apparently!).
Now, when I look at that External HD using my Win 11 Desk-top, it
doesn't 'see' anything.
Is this because Linux is at a 'Higher'/'deeper' level than Windows can
'see'??
I would call that a reasonable observation with reference to the users.
Post by Daniel70
If I back-up this Win-11 Desktop to the same External HD, is there a
possibility that Win-11 will write itself over the Linux Image?? Or is
Win-11 able to detect that there is 'something' there so will go looking
for the next available UNUSED portion of the External HD??
dd if=/dev/sdA of=/somepath/mybackup-2025-03-17.dd bs=16M status=progress

you can also create a partition exactly the same size as the source and
then do

dd if=/dev/sdS of=/dev/sdT bs=16M status=progress


I got "bs=16M" from Carlos some time ago, speeds it up a bit.


If you dd'd a partition you can even boot it and use it but remove the
source drive before you do because both will have the same UUID. You can
also ure a SuperGrub DVD to boot it if you find it difficult. But you
can also mount a backup 'file' and look inside it

mount -o loop /somepath/mybackup-2025-03-17.dd /someotherpath

If you ever boot such a copy you need to be careful with UUID's but also
with /etc/fstab content since other automounted partitions WILL fail to
mount unless the edits are first maid.

It really pays to use removable drive backplane racks, they give you
tons of actual physical control that's increasingly being denied users.
I just got me an icy-dock model that squeezes 6 ssd's + a dvd into a
single 5-1/2 inch external bay and I haven't used ANY fixed internal
drives for 3 decades.
Daniel70
2025-03-17 13:21:20 UTC
Reply
Permalink
Post by bad sector
Post by Daniel70
Some time ago, I backed up my Laptop's 500GB internal HD to an 2TB
external HD using (I forget) ... It might have been 'dd' but I doubt it.
However, the 500GB Internals Image took up 500GB on the External (i.e.
a byte-by-byte image, even the empty bytes, apparently!).
Now, when I look at that External HD using my Win 11 Desk-top, it
doesn't 'see' anything.
Is this because Linux is at a 'Higher'/'deeper' level than Windows can
'see'??
I would call that a reasonable observation with reference to the users.
Post by Daniel70
If I back-up this Win-11 Desktop to the same External HD, is there a
possibility that Win-11 will write itself over the Linux Image?? Or is
Win-11 able to detect that there is 'something' there so will go
looking for the next available UNUSED portion of the External HD??
dd if=/dev/sdA of=/somepath/mybackup-2025-03-17.dd bs=16M status=progress
you can also create a partition exactly the same size as the source and
then do
dd if=/dev/sdS of=/dev/sdT bs=16M status=progress
I got "bs=16M" from Carlos some time ago, speeds it up a bit.
If you dd'd a partition you can even boot it and use it but remove the
source drive before you do because both will have the same UUID. You can
also ure a SuperGrub DVD to boot it if you find it difficult. But you
can also mount a backup 'file' and look inside it
mount -o loop /somepath/mybackup-2025-03-17.dd /someotherpath
If you ever boot such a copy you need to be careful with UUID's but also
with /etc/fstab content since other automounted partitions WILL fail to
mount unless the edits are first maid.
It really pays to use removable drive backplane racks, they give you
tons of actual physical control that's increasingly being denied users.
I just got me an icy-dock model that squeezes 6 ssd's + a dvd into a
single 5-1/2 inch external bay and I haven't used ANY fixed internal
drives for 3 decades.
I have no intention of ever booting from the Back-up, just wanting to
save my data .... but thanks for the suggestions. ;-)
--
Daniel70
Paul
2025-03-17 15:54:50 UTC
Reply
Permalink
Post by bad sector
Some time ago, I backed up my Laptop's 500GB internal HD to an 2TB external HD using (I forget) ... It might have been 'dd' but I doubt it.
However, the 500GB Internals Image took up 500GB on the External (i.e. a byte-by-byte image, even the empty bytes, apparently!).
Now, when I look at that External HD using my Win 11 Desk-top, it doesn't 'see' anything.
Is this because Linux is at a 'Higher'/'deeper' level than Windows can 'see'??
I would call that a reasonable observation with reference to the users.
If I back-up this Win-11 Desktop to the same External HD, is there a possibility that Win-11 will write itself over the Linux Image?? Or is Win-11 able to detect that there is 'something' there so will go looking for the next available UNUSED portion of the External HD??
dd if=/dev/sdA of=/somepath/mybackup-2025-03-17.dd bs=16M status=progress
you can also create a partition exactly the same size as the source and then do
dd if=/dev/sdS of=/dev/sdT bs=16M status=progress
I got "bs=16M" from Carlos some time ago, speeds it up a bit.
If you dd'd a partition you can even boot it and use it but remove the source drive before you do because both will have the same UUID. You can also ure a SuperGrub DVD to boot it if you find it difficult. But you can also mount a backup 'file' and look inside it
mount -o loop /somepath/mybackup-2025-03-17.dd /someotherpath
If you ever boot such a copy you need to be careful with UUID's but also with /etc/fstab content since other automounted partitions WILL fail to mount unless the edits are first maid.
It really pays to use removable drive backplane racks, they give you tons of actual physical control that's increasingly being denied users. I just got me an icy-dock model that squeezes 6 ssd's + a dvd into a single 5-1/2 inch external bay and I haven't used ANY fixed internal drives for 3 decades.
I have no intention of ever booting from the Back-up, just wanting to save my data .... but thanks for the suggestions. ;-)
Only you can do the work, Obiwan.

We can't see what you've done from here.

I can place a 500GB file on a 2TB partition, and I can
analyze it later with "disktype some.file" and see
what is inside the file. If what is inside the file
resembles a hard drive, disktype will burst forth
with a summary of the old disk drive layout.

If there are file systems inside the file, a user can
use a loopback mount, with a byte offset value, and
that will allow reading or writing the partition *inside*
the 500GB file.

If the disk had been "dd" cloned to a second disk, then
things like "gnome-disks" should see the file system sitting
on the 2TB drive. Even disktype can see it (but should not be
needed, as the OS does the analysis for us).

sudo disktype /dev/sdb

You have all the tools you need, at your disposal, to
"determine what is inside a thing".

Paul
Loading...