My Sons Prime Z390-A Asus Windows 10 system i7 8700 CPU will not update
to Windows 11, Windows update says it is incompatible, so I ran
WhyNot11and get 3 red blocks.

Boot method
secure boot
Partition type.

Secure boot seems set up OK in UEFI and TPM is enabled,
But it is installed on an MBR drive, so I decided the first thing to do
was convert that disk to GPT and ran mbr2gpt /validate /allowFullOS
from an elevated cmd prompt.

It goes trough the procedure and comes back with this error
'Validate failed on disk 3'.
That is the proper disk in disk management, Any ideas why?
I have never used , mbr2gpt before so am not sure what to check.

Rene

On 2/11/2022 9:25 PM, Rene Lamontagne wrote:
> My Sons Prime Z390-A Asus Windows 10 system  i7 8700 CPU will not update to Windows 11, Windows update says it is incompatible, so I ran WhyNot11and get 3 red blocks.
>
> Boot method
> secure boot
> Partition type.
>
> Secure boot seems set up OK in UEFI and TPM is enabled,
> But it is installed on an MBR drive, so I decided the first thing to do was convert that disk to GPT and ran   mbr2gpt   /validate /allowFullOS  from an elevated cmd prompt.
>
>  It goes trough the procedure and comes back with this error 'Validate failed on disk 3'.
> That is the proper disk in disk management, Any ideas why?
>  I have never used , mbr2gpt   before so am not sure what to check.
>
> Rene

https://docs.microsoft.com/en-us/windows/deployment/mbr-to-gpt

"Disk Prerequisites

Before any change to the disk is made, MBR2GPT validates the layout and geometry of the selected disk to ensure that:

1 The disk is currently using MBR
2 There is enough space not occupied by partitions to store the primary and secondary GPTs:
16KB + 2 sectors at the front of the disk
16KB + 1 sector at the end of the disk
3 There are at most 3 primary partitions in the MBR partition table
4 One of the partitions is set as active and is the system partition
5 The disk does not have any extended/logical partition
6 The BCD store on the system partition contains a default OS entry pointing to an OS partition
7 The volume IDs can be retrieved for each volume which has a drive letter assigned
8 All partitions on the disk are of MBR types recognized by Windows or has a mapping
specified using the /map command-line option

If any of these checks fails, the conversion will not proceed and an error will be returned.
"

Sounds like the final exam at disk school :-)

For (2), I would scootch the first partition over 1MB, to leave room for primary GPT.
This will require a partition manager of some sort, as moving a partition
to the right, is not on the dining menu of Windows.

The last partition should not be touching the end of the disk, and
normally, it would not be touching anyway. The end of the disk
is a multiple of CHS and is divisible by 63. The last partition
may end on a 1MB quanta, but the last cylinder lays beyond that
and some fraction of ~8MB will remain.

However, the secondary GPT may also suffer from a phobia, and
not like to be lounging at the very very end of the disk. You'll
have to experiment with trimming the end off the last partition,
if it was blatantly touching the end. (This is the easy one,
and can be done from Disk Management "shrink". The shrunk space
would end up where we want it.)

You'll still have to check the rest, as they've taken all the
fun out of it.

Paul

On 2022-02-11 9:07 p.m., Paul wrote:
> On 2/11/2022 9:25 PM, Rene Lamontagne wrote:
>> My Sons Prime Z390-A Asus Windows 10 system  i7 8700 CPU will not
>> update to Windows 11, Windows update says it is incompatible, so I ran
>> WhyNot11and get 3 red blocks.
>>
>> Boot method
>> secure boot
>> Partition type.
>>
>> Secure boot seems set up OK in UEFI and TPM is enabled,
>> But it is installed on an MBR drive, so I decided the first thing to
>> do was convert that disk to GPT and ran   mbr2gpt   /validate
>> /allowFullOS  from an elevated cmd prompt.
>>
>>   It goes trough the procedure and comes back with this error
>> 'Validate failed on disk 3'.
>> That is the proper disk in disk management, Any ideas why?
>>   I have never used , mbr2gpt   before so am not sure what to check.
>>
>> Rene
>
> https://docs.microsoft.com/en-us/windows/deployment/mbr-to-gpt
>
>   "Disk Prerequisites
>
>    Before any change to the disk is made, MBR2GPT validates the layout
> and geometry of the selected disk to ensure that:
>
> 1   The disk is currently using MBR
> 2   There is enough space not occupied by partitions to store the
> primary and secondary GPTs:
>         16KB + 2 sectors at the front of the disk
>         16KB + 1 sector at the end of the disk
> 3   There are at most 3 primary partitions in the MBR partition table
> 4   One of the partitions is set as active and is the system partition
> 5   The disk does not have any extended/logical partition
> 6   The BCD store on the system partition contains a default OS entry
> pointing to an OS partition
> 7   The volume IDs can be retrieved for each volume which has a drive
> letter assigned
> 8   All partitions on the disk are of MBR types recognized by Windows or
> has a mapping
>     specified using the /map command-line option
>
>     If any of these checks fails, the conversion will not proceed and
> an error will be returned.
>    "
>
> Sounds like the final exam at disk school :-)
>
> For (2), I would scootch the first partition over 1MB, to leave room for
> primary GPT.
> This will require a partition manager of some sort, as moving a partition
> to the right, is not on the dining menu of Windows.
>
> The last partition should not be touching the end of the disk, and
> normally, it would not be touching anyway. The end of the disk
> is a multiple of CHS and is divisible by 63. The last partition
> may end on a 1MB quanta, but the last cylinder lays beyond that
> and some fraction of ~8MB will remain.
>
> However, the secondary GPT may also suffer from a phobia, and
> not like to be lounging at the very very end of the disk. You'll
> have to experiment with trimming the end off the last partition,
> if it was blatantly touching the end. (This is the easy one,
> and can be done from Disk Management "shrink". The shrunk space
> would end up where we want it.)
>
> You'll still have to check the rest, as they've taken all the
> fun out of it.
>
>    Paul
>
>

Good stuff Paul, I will have access to the computer on Sunday and then
maybe sort it out, I am sure there are 4 or 5 partition and no free
space so will start there, Then i will reset to make sure it boots UEFI
and will verify secure boot also. Will report back next week.

Rene

On 2/12/22 04:55, Rene Lamontagne wrote:
> On 2022-02-11 9:07 p.m., Paul wrote:
>> On 2/11/2022 9:25 PM, Rene Lamontagne wrote:
>>> My Sons Prime Z390-A Asus Windows 10 system  i7 8700 CPU will not
>>> update to Windows 11, Windows update says it is incompatible, so I
>>> ran WhyNot11and get 3 red blocks.
>>>
>>> Boot method
>>> secure boot
>>> Partition type.
>>>
>>> Secure boot seems set up OK in UEFI and TPM is enabled,
>>> But it is installed on an MBR drive, so I decided the first thing to
>>> do was convert that disk to GPT and ran   mbr2gpt   /validate
>>> /allowFullOS  from an elevated cmd prompt.
>>>
>>>   It goes trough the procedure and comes back with this error
>>> 'Validate failed on disk 3'.
>>> That is the proper disk in disk management, Any ideas why?
>>>   I have never used , mbr2gpt   before so am not sure what to check.
>>>
>>> Rene
>>
>> https://docs.microsoft.com/en-us/windows/deployment/mbr-to-gpt
>>
>>    "Disk Prerequisites
>>
>>     Before any change to the disk is made, MBR2GPT validates the
>> layout and geometry of the selected disk to ensure that:
>>
>> 1   The disk is currently using MBR
>> 2   There is enough space not occupied by partitions to store the
>> primary and secondary GPTs:
>>          16KB + 2 sectors at the front of the disk
>>          16KB + 1 sector at the end of the disk
>> 3   There are at most 3 primary partitions in the MBR partition table
>> 4   One of the partitions is set as active and is the system partition
>> 5   The disk does not have any extended/logical partition
>> 6   The BCD store on the system partition contains a default OS entry
>> pointing to an OS partition
>> 7   The volume IDs can be retrieved for each volume which has a drive
>> letter assigned
>> 8   All partitions on the disk are of MBR types recognized by Windows
>> or has a mapping
>>      specified using the /map command-line option
>>
>>      If any of these checks fails, the conversion will not proceed and
>> an error will be returned.
>>     "
>>
>> Sounds like the final exam at disk school :-)
>>
>> For (2), I would scootch the first partition over 1MB, to leave room
>> for primary GPT.
>> This will require a partition manager of some sort, as moving a partition
>> to the right, is not on the dining menu of Windows.
>>
>> The last partition should not be touching the end of the disk, and
>> normally, it would not be touching anyway. The end of the disk
>> is a multiple of CHS and is divisible by 63. The last partition
>> may end on a 1MB quanta, but the last cylinder lays beyond that
>> and some fraction of ~8MB will remain.
>>
>> However, the secondary GPT may also suffer from a phobia, and
>> not like to be lounging at the very very end of the disk. You'll
>> have to experiment with trimming the end off the last partition,
>> if it was blatantly touching the end. (This is the easy one,
>> and can be done from Disk Management "shrink". The shrunk space
>> would end up where we want it.)
>>
>> You'll still have to check the rest, as they've taken all the
>> fun out of it.
>>
>>     Paul
>>
>>
>
> Good stuff Paul, I will have access to the computer on Sunday and then
> maybe sort it out, I am sure there are 4 or 5 partition and no free
> space so will start there, Then i will reset to make sure it boots UEFI
> and will verify secure boot also. Will report back next week.
>
> Rene
>
>
>
Don't forget to backup, if I were you I would certainly not try it on
the current disk but try it on some old disk I have lying around first.

On 2022-02-11 9:55 p.m., Rene Lamontagne wrote:
> On 2022-02-11 9:07 p.m., Paul wrote:
>> On 2/11/2022 9:25 PM, Rene Lamontagne wrote:
>>> My Sons Prime Z390-A Asus Windows 10 system  i7 8700 CPU will not
>>> update to Windows 11, Windows update says it is incompatible, so I
>>> ran WhyNot11and get 3 red blocks.
>>>
>>> Boot method
>>> secure boot
>>> Partition type.
>>>
>>> Secure boot seems set up OK in UEFI and TPM is enabled,
>>> But it is installed on an MBR drive, so I decided the first thing to
>>> do was convert that disk to GPT and ran   mbr2gpt   /validate
>>> /allowFullOS  from an elevated cmd prompt.
>>>
>>>   It goes trough the procedure and comes back with this error
>>> 'Validate failed on disk 3'.
>>> That is the proper disk in disk management, Any ideas why?
>>>   I have never used , mbr2gpt   before so am not sure what to check.
>>>
>>> Rene
>>
>> https://docs.microsoft.com/en-us/windows/deployment/mbr-to-gpt
>>
>>    "Disk Prerequisites
>>
>>     Before any change to the disk is made, MBR2GPT validates the
>> layout and geometry of the selected disk to ensure that:
>>
>> 1   The disk is currently using MBR
>> 2   There is enough space not occupied by partitions to store the
>> primary and secondary GPTs:
>>          16KB + 2 sectors at the front of the disk
>>          16KB + 1 sector at the end of the disk
>> 3   There are at most 3 primary partitions in the MBR partition table
>> 4   One of the partitions is set as active and is the system partition
>> 5   The disk does not have any extended/logical partition
>> 6   The BCD store on the system partition contains a default OS entry
>> pointing to an OS partition
>> 7   The volume IDs can be retrieved for each volume which has a drive
>> letter assigned
>> 8   All partitions on the disk are of MBR types recognized by Windows
>> or has a mapping
>>      specified using the /map command-line option
>>
>>      If any of these checks fails, the conversion will not proceed and
>> an error will be returned.
>>     "
>>
>> Sounds like the final exam at disk school :-)
>>
>> For (2), I would scootch the first partition over 1MB, to leave room
>> for primary GPT.
>> This will require a partition manager of some sort, as moving a partition
>> to the right, is not on the dining menu of Windows.
>>
>> The last partition should not be touching the end of the disk, and
>> normally, it would not be touching anyway. The end of the disk
>> is a multiple of CHS and is divisible by 63. The last partition
>> may end on a 1MB quanta, but the last cylinder lays beyond that
>> and some fraction of ~8MB will remain.
>>
>> However, the secondary GPT may also suffer from a phobia, and
>> not like to be lounging at the very very end of the disk. You'll
>> have to experiment with trimming the end off the last partition,
>> if it was blatantly touching the end. (This is the easy one,
>> and can be done from Disk Management "shrink". The shrunk space
>> would end up where we want it.)
>>
>> You'll still have to check the rest, as they've taken all the
>> fun out of it.
>>
>>     Paul
>>
>>
>
> Good stuff Paul, I will have access to the computer on Sunday and then
> maybe sort it out, I am sure there are 4 or 5 partition and no free
> space so will start there, Then i will reset to make sure it boots UEFI
> and will verify secure boot also. Will report back next week.
>
> Rene
>
>
>

By the way, I will be using Minitool Partition Wizard as I find it full
featured and very flexible. much better than Windows disk management.

Rene

Rene

On 2022-02-11 10:13 p.m., dillinger wrote:
> On 2/12/22 04:55, Rene Lamontagne wrote:
>
>> On 2022-02-11 9:07 p.m., Paul wrote:
>>> On 2/11/2022 9:25 PM, Rene Lamontagne wrote:
>>>> My Sons Prime Z390-A Asus Windows 10 system  i7 8700 CPU will not
>>>> update to Windows 11, Windows update says it is incompatible, so I
>>>> ran WhyNot11and get 3 red blocks.
>>>>
>>>> Boot method
>>>> secure boot
>>>> Partition type.
>>>>
>>>> Secure boot seems set up OK in UEFI and TPM is enabled,
>>>> But it is installed on an MBR drive, so I decided the first thing to
>>>> do was convert that disk to GPT and ran   mbr2gpt   /validate
>>>> /allowFullOS  from an elevated cmd prompt.
>>>>
>>>>   It goes trough the procedure and comes back with this error
>>>> 'Validate failed on disk 3'.
>>>> That is the proper disk in disk management, Any ideas why?
>>>>   I have never used , mbr2gpt   before so am not sure what to check.
>>>>
>>>> Rene
>>>
>>> https://docs.microsoft.com/en-us/windows/deployment/mbr-to-gpt
>>>
>>>    "Disk Prerequisites
>>>
>>>     Before any change to the disk is made, MBR2GPT validates the
>>> layout and geometry of the selected disk to ensure that:
>>>
>>> 1   The disk is currently using MBR
>>> 2   There is enough space not occupied by partitions to store the
>>> primary and secondary GPTs:
>>>          16KB + 2 sectors at the front of the disk
>>>          16KB + 1 sector at the end of the disk
>>> 3   There are at most 3 primary partitions in the MBR partition table
>>> 4   One of the partitions is set as active and is the system partition
>>> 5   The disk does not have any extended/logical partition
>>> 6   The BCD store on the system partition contains a default OS entry
>>> pointing to an OS partition
>>> 7   The volume IDs can be retrieved for each volume which has a drive
>>> letter assigned
>>> 8   All partitions on the disk are of MBR types recognized by Windows
>>> or has a mapping
>>>      specified using the /map command-line option
>>>
>>>      If any of these checks fails, the conversion will not proceed
>>> and an error will be returned.
>>>     "
>>>
>>> Sounds like the final exam at disk school :-)
>>>
>>> For (2), I would scootch the first partition over 1MB, to leave room
>>> for primary GPT.
>>> This will require a partition manager of some sort, as moving a
>>> partition
>>> to the right, is not on the dining menu of Windows.
>>>
>>> The last partition should not be touching the end of the disk, and
>>> normally, it would not be touching anyway. The end of the disk
>>> is a multiple of CHS and is divisible by 63. The last partition
>>> may end on a 1MB quanta, but the last cylinder lays beyond that
>>> and some fraction of ~8MB will remain.
>>>
>>> However, the secondary GPT may also suffer from a phobia, and
>>> not like to be lounging at the very very end of the disk. You'll
>>> have to experiment with trimming the end off the last partition,
>>> if it was blatantly touching the end. (This is the easy one,
>>> and can be done from Disk Management "shrink". The shrunk space
>>> would end up where we want it.)
>>>
>>> You'll still have to check the rest, as they've taken all the
>>> fun out of it.
>>>
>>>     Paul
>>>
>>>
>>
>> Good stuff Paul, I will have access to the computer on Sunday and then
>> maybe sort it out, I am sure there are 4 or 5 partition and no free
>> space so will start there, Then i will reset to make sure it boots
>> UEFI and will verify secure boot also. Will report back next week.
>>
>> Rene
>>
>>
>>
> Don't forget to backup, if I were you I would certainly not try it on
> the current disk but try it on some old disk I have lying around first.

I Always do Macrium Reflect backups before messing with any drive.
I tried it on a scratch WD 160GB drive and on a 256GB Samsung SSD and it
worked fine.

Also now I find that I can do this directly from 'MiniTool Partition
Wizard',
and I can set it back to MBR, so should be good to go tomorrow.

Rene

On 2022-02-11 10:19 p.m., Rene Lamontagne wrote:
> On 2022-02-11 9:55 p.m., Rene Lamontagne wrote:
>> On 2022-02-11 9:07 p.m., Paul wrote:
>>> On 2/11/2022 9:25 PM, Rene Lamontagne wrote:
>>>> My Sons Prime Z390-A Asus Windows 10 system  i7 8700 CPU will not
>>>> update to Windows 11, Windows update says it is incompatible, so I
>>>> ran WhyNot11and get 3 red blocks.
>>>>
>>>> Boot method
>>>> secure boot
>>>> Partition type.
>>>>
>>>> Secure boot seems set up OK in UEFI and TPM is enabled,
>>>> But it is installed on an MBR drive, so I decided the first thing to
>>>> do was convert that disk to GPT and ran   mbr2gpt   /validate
>>>> /allowFullOS  from an elevated cmd prompt.
>>>>
>>>>   It goes trough the procedure and comes back with this error
>>>> 'Validate failed on disk 3'.
>>>> That is the proper disk in disk management, Any ideas why?
>>>>   I have never used , mbr2gpt   before so am not sure what to check.
>>>>
>>>> Rene
>>>
>>> https://docs.microsoft.com/en-us/windows/deployment/mbr-to-gpt
>>>
>>>    "Disk Prerequisites
>>>
>>>     Before any change to the disk is made, MBR2GPT validates the
>>> layout and geometry of the selected disk to ensure that:
>>>
>>> 1   The disk is currently using MBR
>>> 2   There is enough space not occupied by partitions to store the
>>> primary and secondary GPTs:
>>>          16KB + 2 sectors at the front of the disk
>>>          16KB + 1 sector at the end of the disk
>>> 3   There are at most 3 primary partitions in the MBR partition table
>>> 4   One of the partitions is set as active and is the system partition
>>> 5   The disk does not have any extended/logical partition
>>> 6   The BCD store on the system partition contains a default OS entry
>>> pointing to an OS partition
>>> 7   The volume IDs can be retrieved for each volume which has a drive
>>> letter assigned
>>> 8   All partitions on the disk are of MBR types recognized by Windows
>>> or has a mapping
>>>      specified using the /map command-line option
>>>
>>>      If any of these checks fails, the conversion will not proceed
>>> and an error will be returned.
>>>     "
>>>
>>> Sounds like the final exam at disk school :-)
>>>
>>> For (2), I would scootch the first partition over 1MB, to leave room
>>> for primary GPT.
>>> This will require a partition manager of some sort, as moving a
>>> partition
>>> to the right, is not on the dining menu of Windows.
>>>
>>> The last partition should not be touching the end of the disk, and
>>> normally, it would not be touching anyway. The end of the disk
>>> is a multiple of CHS and is divisible by 63. The last partition
>>> may end on a 1MB quanta, but the last cylinder lays beyond that
>>> and some fraction of ~8MB will remain.
>>>
>>> However, the secondary GPT may also suffer from a phobia, and
>>> not like to be lounging at the very very end of the disk. You'll
>>> have to experiment with trimming the end off the last partition,
>>> if it was blatantly touching the end. (This is the easy one,
>>> and can be done from Disk Management "shrink". The shrunk space
>>> would end up where we want it.)
>>>
>>> You'll still have to check the rest, as they've taken all the
>>> fun out of it.
>>>
>>>     Paul
>>>
>>>
>>
>> Good stuff Paul, I will have access to the computer on Sunday and then
>> maybe sort it out, I am sure there are 4 or 5 partition and no free
>> space so will start there, Then i will reset to make sure it boots
>> UEFI and will verify secure boot also. Will report back next week.
>>
>> Rene
>>
>>
>>
>
> By the way, I will be using Minitool Partition Wizard as I find it full
> featured and very  flexible. much better than Windows disk management.
>
> Rene
>
>
> Rene

Well got some time with his system today, here is where it was, neither
elevated command line or Minitool Partition Wizard could validate the drive.

It had 5 partitions so I deleted 2 which left 300GB of unallocated
space, tried again , no dice.

I then moved C: over to the right to make room for the 8GB blank spaces
both ahead and after C:, This resulted in a not able to boot condition,
Had to load Macrium reflect rescue disk and run the Boot repair module
which brought Windows back to life.

This led to mbr2gpt throwing this message 'Cannot find OS partition C:'

So I found this site which gave me a ton of good info on MBR and GPT
problems.

https://www.partitionwizard.com/partitionmagic/fix-mbr2gpt-failed-errors.html

In there I found this error fix for that error,

bcdboot C:\ windows/ F bios/S C:
which I applied, This then allowed mbr2gpt to give me a 'validate
successful' on the drive

So now tried again from the command line to "convert' , it started and
did part of it and got this error message

Not enough room to create the EFI partition, I have about 65GB free on
C: and 300GB of unallocated room on the drive so I am thinking this
space needs to be before C:\

My Son then needed the use of the machine so will continue next week and
create another unallocated 8GB in front of C: maybe that will do it.

Rene

On 2/13/2022 6:09 PM, Rene Lamontagne wrote:
> On 2022-02-11 10:19 p.m., Rene Lamontagne wrote:
>> On 2022-02-11 9:55 p.m., Rene Lamontagne wrote:
>>> On 2022-02-11 9:07 p.m., Paul wrote:
>>>> On 2/11/2022 9:25 PM, Rene Lamontagne wrote:
>>>>> My Sons Prime Z390-A Asus Windows 10 system  i7 8700 CPU will not update to Windows 11, Windows update says it is incompatible, so I ran WhyNot11and get 3 red blocks.
>>>>>
>>>>> Boot method
>>>>> secure boot
>>>>> Partition type.
>>>>>
>>>>> Secure boot seems set up OK in UEFI and TPM is enabled,
>>>>> But it is installed on an MBR drive, so I decided the first thing to do was convert that disk to GPT and ran   mbr2gpt   /validate /allowFullOS  from an elevated cmd prompt.
>>>>>
>>>>>   It goes trough the procedure and comes back with this error 'Validate failed on disk 3'.
>>>>> That is the proper disk in disk management, Any ideas why?
>>>>>   I have never used , mbr2gpt   before so am not sure what to check.
>>>>>
>>>>> Rene
>>>>
>>>> https://docs.microsoft.com/en-us/windows/deployment/mbr-to-gpt
>>>>
>>>>    "Disk Prerequisites
>>>>
>>>>     Before any change to the disk is made, MBR2GPT validates the layout and geometry of the selected disk to ensure that:
>>>>
>>>> 1   The disk is currently using MBR
>>>> 2   There is enough space not occupied by partitions to store the primary and secondary GPTs:
>>>>          16KB + 2 sectors at the front of the disk
>>>>          16KB + 1 sector at the end of the disk
>>>> 3   There are at most 3 primary partitions in the MBR partition table
>>>> 4   One of the partitions is set as active and is the system partition
>>>> 5   The disk does not have any extended/logical partition
>>>> 6   The BCD store on the system partition contains a default OS entry pointing to an OS partition
>>>> 7   The volume IDs can be retrieved for each volume which has a drive letter assigned
>>>> 8   All partitions on the disk are of MBR types recognized by Windows or has a mapping
>>>>      specified using the /map command-line option
>>>>
>>>>      If any of these checks fails, the conversion will not proceed and an error will be returned.
>>>>     "
>>>>
>>>> Sounds like the final exam at disk school :-)
>>>>
>>>> For (2), I would scootch the first partition over 1MB, to leave room for primary GPT.
>>>> This will require a partition manager of some sort, as moving a partition
>>>> to the right, is not on the dining menu of Windows.
>>>>
>>>> The last partition should not be touching the end of the disk, and
>>>> normally, it would not be touching anyway. The end of the disk
>>>> is a multiple of CHS and is divisible by 63. The last partition
>>>> may end on a 1MB quanta, but the last cylinder lays beyond that
>>>> and some fraction of ~8MB will remain.
>>>>
>>>> However, the secondary GPT may also suffer from a phobia, and
>>>> not like to be lounging at the very very end of the disk. You'll
>>>> have to experiment with trimming the end off the last partition,
>>>> if it was blatantly touching the end. (This is the easy one,
>>>> and can be done from Disk Management "shrink". The shrunk space
>>>> would end up where we want it.)
>>>>
>>>> You'll still have to check the rest, as they've taken all the
>>>> fun out of it.
>>>>
>>>>     Paul
>>>>
>>>>
>>>
>>> Good stuff Paul, I will have access to the computer on Sunday and then maybe sort it out, I am sure there are 4 or 5 partition and no free space so will start there, Then i will reset to make sure it boots UEFI and will verify secure boot also. Will report back next week.
>>>
>>> Rene
>>>
>>>
>>>
>>
>> By the way, I will be using Minitool Partition Wizard as I find it full featured and very  flexible. much better than Windows disk management.
>>
>> Rene
>>
>>
>> Rene
>
>
> Well got some time with his system today, here is where it was,  neither elevated command line or Minitool Partition Wizard could validate the drive.
>
> It had 5 partitions so I deleted 2 which left 300GB of unallocated space, tried again , no dice.
>
> I then moved C: over to the right to make room for the 8GB blank spaces both ahead and after C:, This resulted in a not able to boot condition, Had to load Macrium reflect rescue disk and run  the Boot repair module which brought Windows back to life.
>
> This led to mbr2gpt throwing this message 'Cannot find OS partition C:'
>
> So I found this site which gave me a ton of good info on MBR and GPT problems.
>
> https://www.partitionwizard.com/partitionmagic/fix-mbr2gpt-failed-errors.html
>
> In there I found this error fix for that error,
>
> bcdboot C:\ windows/ F bios/S C:
>  which I applied, This then allowed mbr2gpt to give me a 'validate successful' on the drive
>
> So now tried again from the command line to "convert' , it started and did part of it and got this error message
>
> Not enough room to create the EFI partition, I have about 65GB free on C: and 300GB of unallocated room on the drive so I am thinking this space needs to be before C:\
>
> My Son then needed the use of the machine so will continue next week and create another unallocated 8GB in front of C: maybe that will do it.
>
> Rene

I'm making modifications right now, to a multiboot GPT disk,
and all I've been able to conclude so far, is "it's complicated".
Different tools seem to have different disciplines in terms of
partition numbering. Some utilities just scramble the things.

In the old days, a basic check, was to take PTEDIT32 and examine
the number of the partition, and its origin. Only accept disk
setups, that have them in monotonically increasing order. Because
to do otherwise, will "set you up for a fall" later. You always
want to clean up your work area, remove the oily rags, before
starting the next project :-)

When I used to use PowerQuest, I would use PTEDIT32.exe later,
and swap lines in the partition table, then go into boot.ini
and edit the ARC and correct it. This is something the staff
at PowerQuest could have done in their code. But being chickenshits,
they did not. Leaving me to do this manually. They would make
slot4 point to Partition3 and slot3 point to Partition4. You can't
do that, without losing your mind later, when bad things happen
(TestDisk).

Sure, we use tools every day, never knowing what kind of a mess
they're making underneath.

I tried to edit an old Mac disk one day, using gparted. Gparted
told me, "oh, YES, I support this format". Well, after clicking
the Apply button, the disk was completely destroyed. But, I had
a "dd" backup, so you cannot put me off the trail that easily.
Examining the partition table in a Hex Editor later, it wasn't
even apparent it understood the partition table. As each partition
had been destroyed a different way.

At least you had your backup, and like me, you're "dialing in" the
tool, to see what makes it tick, and what makes it explode.

Paul

Rene Lamontagne wrote:
>
> Not enough room to create the EFI partition, I have about 65GB free on
> C: and 300GB of unallocated room on the drive so I am thinking this
> space needs to be before C:\
>
> My Son then needed the use of the machine so will continue next week and
> create another unallocated 8GB in front of C: maybe that will do it.
>
> Rene
>
>
GPT partiton order on disk
1st EFI
2nd MSR
3rd Windows
4th Recovery
5th and later - anything else

EFI 100 MB
MSR 16 MB
Windows - optional
Recovery 500 MB(1 GB is better)

--
....w¡ñ§±¤ñ

On 2022-02-14 1:39 a.m., ...w¡ñ§±¤ñ wrote:
> Rene Lamontagne wrote:
>>
>> Not enough room to create the EFI partition, I have about 65GB free on
>> C: and 300GB of unallocated room on the drive so I am thinking this
>> space needs to be before C:\
>>
>> My Son then needed the use of the machine so will continue next week
>> and create another unallocated 8GB in front of C: maybe that will do it.
>>
>> Rene
>>
>>
> GPT partiton order on disk
> 1st EFI
> 2nd MSR
> 3rd Windows
> 4th Recovery
> 5th and later - anything else
>
> EFI 100 MB
> MSR 16 MB
> Windows - optional
> Recovery 500 MB(1 GB is better)
>
>

Yes some were missing so tried the mbr2gpt thing again today, Now it
says 'operation failed, could not shrink Windcows partion C,'

Now why in hell did it want to do that? C: is 100GB and there is over
350GB of unallocated space on the disk, tried again, same results

Well to see what would happen I resized Windows to 120GB and tried again,
A minor miracle happened and mbr2gpt came back with 'Operation successful'.

Now that drive shows up as GPT at last, Next I ran Macrium reflect
'Repair windows boot problems and that brought back all the missing
partitions on the disk.

Whynot11 now shows all 11 blocks Green and Windows update shows computer
is compatible with Windows 11, So Sunday I will install 11, Well, That
that was a bit of a struggle, Thanks to all for your help.

Rene

Rene Lamontagne wrote:
> On 2022-02-14 1:39 a.m., ...w¡ñ§±¤ñ wrote:
>> Rene Lamontagne wrote:
>>>
>>> Not enough room to create the EFI partition, I have about 65GB free
>>> on C: and 300GB of unallocated room on the drive so I am thinking
>>> this space needs to be before C:\
>>>
>>> My Son then needed the use of the machine so will continue next week
>>> and create another unallocated 8GB in front of C: maybe that will do it.
>>>
>>> Rene
>>>
>>>
>> GPT partiton order on disk
>> 1st EFI
>> 2nd MSR
>> 3rd Windows
>> 4th Recovery
>> 5th and later - anything else
>>
>> EFI 100 MB
>> MSR 16 MB
>> Windows - optional
>> Recovery 500 MB(1 GB is better)
>>
>>
>
> Yes some were missing so tried the mbr2gpt  thing again today, Now it
> says 'operation failed, could not shrink Windcows partion C,'
>
> Now why in hell did it want to do that?  C: is 100GB and there is over
> 350GB of unallocated space on the disk, tried again, same results
>
> Well to see what would happen I resized Windows to 120GB and tried again,
> A minor miracle happened and mbr2gpt came back with  'Operation
> successful'.
>
> Now that drive shows up as GPT at last, Next I ran Macrium reflect
> 'Repair windows boot problems and that brought back all the missing
> partitions on the disk.
>
> Whynot11 now shows all 11 blocks Green and Windows update shows computer
> is compatible with Windows 11, So Sunday I will install 11, Well, That
> that was a bit of a struggle, Thanks to all for your help.
>
> Rene
>
>
>
>
Progess. Congratulations.
Revisit my post on the partition order for GPT.
- that order is the usual(and recommended) order.

Looking back at your earlier error messages, in hindsight it may have
helped to include or link to a picture of the partitions on the disk
which showed their position and size.

Even now it still might be appropriate to see a picture or the end
result to determine if those GPT partitions are in the recommended order.
Why..in some cases after conversion the order isn't exactly as
recommended. MSFT fixed a problem with the Recovery being created at the
beginning of the disk and the EFI after the Windows which is contrary to
the recommended GPT order(System-EFI, MSR, Windows, Recovery).
- With the Recovery at the beginning, if the size of the Recovery is
insufficient, it can not be resized larger thus an additional and the
new current Recovery will be created in the recommended position(after
the Windows) by shrinking the Windows partition - typically done if
necessary with a feature update or o/s change(e.g. later 10
version/build and/or 10 to 11 o/s upgrade). Second, if a new EFI is
created elswhere on the disk(after the Windows) then the end result
could be a disk with a no longer user older System partiton(remnant of
MBR at the beginning of the disk) and an extra an no longer
used/necessary Recovery partition(at beginning or end of disk.

With the partitions in the recommended order(System, MSR, Windows,
Recovery - whenever the Recovery needs more space, by design Windows
will shrink the Windows allowing the Recovery(adjacent and to right of
Windows parition) to be resized larger preventing the recreation of a
new Recovery partition.

If you've time - Post or link to a picture or elaborate further on the
partitions and order.

--
....w¡ñ§±¤ñ

On 2022-02-23 3:30 p.m., ...w¡ñ§±¤ñ wrote:
> Rene Lamontagne wrote:
>> On 2022-02-14 1:39 a.m., ...w¡ñ§±¤ñ wrote:
>>> Rene Lamontagne wrote:
>>>>
>>>> Not enough room to create the EFI partition, I have about 65GB free
>>>> on C: and 300GB of unallocated room on the drive so I am thinking
>>>> this space needs to be before C:\
>>>>
>>>> My Son then needed the use of the machine so will continue next week
>>>> and create another unallocated 8GB in front of C: maybe that will do
>>>> it.
>>>>
>>>> Rene
>>>>
>>>>
>>> GPT partiton order on disk
>>> 1st EFI
>>> 2nd MSR
>>> 3rd Windows
>>> 4th Recovery
>>> 5th and later - anything else
>>>
>>> EFI 100 MB
>>> MSR 16 MB
>>> Windows - optional
>>> Recovery 500 MB(1 GB is better)
>>>
>>>
>>
>> Yes some were missing so tried the mbr2gpt  thing again today, Now it
>> says 'operation failed, could not shrink Windcows partion C,'
>>
>> Now why in hell did it want to do that?  C: is 100GB and there is over
>> 350GB of unallocated space on the disk, tried again, same results
>>
>> Well to see what would happen I resized Windows to 120GB and tried again,
>> A minor miracle happened and mbr2gpt came back with  'Operation
>> successful'.
>>
>> Now that drive shows up as GPT at last, Next I ran Macrium reflect
>> 'Repair windows boot problems and that brought back all the missing
>> partitions on the disk.
>>
>> Whynot11 now shows all 11 blocks Green and Windows update shows
>> computer is compatible with Windows 11, So Sunday I will install 11,
>> Well, That that was a bit of a struggle, Thanks to all for your help.
>>
>> Rene
>>
>>
>>
>>
> Progess. Congratulations.
>  Revisit my post on the partition order for GPT.
>  - that order is the usual(and recommended) order.
>
> Looking back at your earlier error messages, in hindsight it may have
> helped to include or link to a picture of the partitions on the disk
> which showed their position and size.
>
> Even now it still might be appropriate to see a picture or the end
> result to determine if those GPT partitions are in the recommended order.
>  Why..in some cases after conversion the order isn't exactly as
> recommended. MSFT fixed a problem with the Recovery being created at the
> beginning of the disk and the EFI after the Windows which is contrary to
> the recommended GPT order(System-EFI, MSR, Windows, Recovery).
> - With the Recovery at the beginning, if the size of the Recovery is
> insufficient, it can not be resized larger thus an additional and the
> new current Recovery will be created in the recommended position(after
> the Windows) by shrinking the Windows partition - typically done if
> necessary with a feature update or o/s change(e.g. later 10
> version/build and/or 10 to 11 o/s upgrade). Second, if a new EFI is
> created elswhere on the disk(after the Windows) then the end result
> could be a disk with a no longer user older System partiton(remnant of
> MBR at the beginning of the disk) and an extra an no longer
> used/necessary Recovery partition(at beginning or end of disk.
>
> With the partitions in the recommended order(System, MSR, Windows,
> Recovery - whenever the Recovery needs more space, by design Windows
> will shrink the Windows allowing the Recovery(adjacent and to right of
> Windows parition) to be resized larger preventing the recreation of a
> new Recovery partition.
>
> If you've time - Post or link to a picture or elaborate further on the
> partitions and order.
>
>
OK as i rememberit .can'yt accss till next week

1. 8GB unallocated
2. 16GB other
3. 100MB EFI
4. 616MB recovery
5. other data partitions

I think that is fairly accurate, The 8GB came about when I moved Windows
over, so I think i will leave it alone, I don't need the space.

Rene

On 2022-02-23 10:42 p.m., Rene Lamontagne wrote:
> On 2022-02-23 3:30 p.m., ...w¡ñ§±¤ñ wrote:
>> Rene Lamontagne wrote:
>>> On 2022-02-14 1:39 a.m., ...w¡ñ§±¤ñ wrote:
>>>> Rene Lamontagne wrote:
>>>>>
>>>>> Not enough room to create the EFI partition, I have about 65GB free
>>>>> on C: and 300GB of unallocated room on the drive so I am thinking
>>>>> this space needs to be before C:\
>>>>>
>>>>> My Son then needed the use of the machine so will continue next
>>>>> week and create another unallocated 8GB in front of C: maybe that
>>>>> will do it.
>>>>>
>>>>> Rene
>>>>>
>>>>>
>>>> GPT partiton order on disk
>>>> 1st EFI
>>>> 2nd MSR
>>>> 3rd Windows
>>>> 4th Recovery
>>>> 5th and later - anything else
>>>>
>>>> EFI 100 MB
>>>> MSR 16 MB
>>>> Windows - optional
>>>> Recovery 500 MB(1 GB is better)
>>>>
>>>>
>>>
>>> Yes some were missing so tried the mbr2gpt  thing again today, Now it
>>> says 'operation failed, could not shrink Windcows partion C,'
>>>
>>> Now why in hell did it want to do that?  C: is 100GB and there is
>>> over 350GB of unallocated space on the disk, tried again, same results
>>>
>>> Well to see what would happen I resized Windows to 120GB and tried
>>> again,
>>> A minor miracle happened and mbr2gpt came back with  'Operation
>>> successful'.
>>>
>>> Now that drive shows up as GPT at last, Next I ran Macrium reflect
>>> 'Repair windows boot problems and that brought back all the missing
>>> partitions on the disk.
>>>
>>> Whynot11 now shows all 11 blocks Green and Windows update shows
>>> computer is compatible with Windows 11, So Sunday I will install 11,
>>> Well, That that was a bit of a struggle, Thanks to all for your help.
>>>
>>> Rene
>>>
>>>
>>>
>>>
>> Progess. Congratulations.
>>   Revisit my post on the partition order for GPT.
>>   - that order is the usual(and recommended) order.
>>
>> Looking back at your earlier error messages, in hindsight it may have
>> helped to include or link to a picture of the partitions on the disk
>> which showed their position and size.
>>
>> Even now it still might be appropriate to see a picture or the end
>> result to determine if those GPT partitions are in the recommended order.
>>   Why..in some cases after conversion the order isn't exactly as
>> recommended. MSFT fixed a problem with the Recovery being created at
>> the beginning of the disk and the EFI after the Windows which is
>> contrary to the recommended GPT order(System-EFI, MSR, Windows,
>> Recovery).
>> - With the Recovery at the beginning, if the size of the Recovery is
>> insufficient, it can not be resized larger thus an additional and the
>> new current Recovery will be created in the recommended position(after
>> the Windows) by shrinking the Windows partition - typically done if
>> necessary with a feature update or o/s change(e.g. later 10
>> version/build and/or 10 to 11 o/s upgrade). Second, if a new EFI is
>> created elswhere on the disk(after the Windows) then the end result
>> could be a disk with a no longer user older System partiton(remnant of
>> MBR at the beginning of the disk) and an extra an no longer
>> used/necessary Recovery partition(at beginning or end of disk.
>>
>> With the partitions in the recommended order(System, MSR, Windows,
>> Recovery - whenever the Recovery needs more space, by design Windows
>> will shrink the Windows allowing the Recovery(adjacent and to right of
>> Windows parition) to be resized larger preventing the recreation of a
>> new Recovery partition.
>>
>> If you've time - Post or link to a picture or elaborate further on the
>> partitions and order.
>>
>>
> OK as i rememberit .can'yt accss till next week
>
> 1.  8GB unallocated
> 2. 16GB other
> 3. 100MB EFI

4. Windows C: 120GB

> 5. 616MB recovery
> 6. other data partitions
>
> I think that is fairly accurate, The 8GB came about when I moved Windows
> over, so I think i will leave it alone,  I don't need the space.
>
> Rene

> Screwed that up, Windows should have been 4.

Rene

Rene Lamontagne wrote:
> On 2022-02-23 10:42 p.m., Rene Lamontagne wrote:
>> On 2022-02-23 3:30 p.m., ...w¡ñ§±¤ñ wrote:
>>> Rene Lamontagne wrote:
>>>> On 2022-02-14 1:39 a.m., ...w¡ñ§±¤ñ wrote:
>>>>> Rene Lamontagne wrote:
>>>>>>
>>>>>> Not enough room to create the EFI partition, I have about 65GB
>>>>>> free on C: and 300GB of unallocated room on the drive so I am
>>>>>> thinking this space needs to be before C:\
>>>>>>
>>>>>> My Son then needed the use of the machine so will continue next
>>>>>> week and create another unallocated 8GB in front of C: maybe that
>>>>>> will do it.
>>>>>>
>>>>>> Rene
>>>>>>
>>>>>>
>>>>> GPT partiton order on disk
>>>>> 1st EFI
>>>>> 2nd MSR
>>>>> 3rd Windows
>>>>> 4th Recovery
>>>>> 5th and later - anything else
>>>>>
>>>>> EFI 100 MB
>>>>> MSR 16 MB
>>>>> Windows - optional
>>>>> Recovery 500 MB(1 GB is better)
>>>>>
>>>>>
>>>>
>>>> Yes some were missing so tried the mbr2gpt  thing again today, Now
>>>> it says 'operation failed, could not shrink Windcows partion C,'
>>>>
>>>> Now why in hell did it want to do that?  C: is 100GB and there is
>>>> over 350GB of unallocated space on the disk, tried again, same results
>>>>
>>>> Well to see what would happen I resized Windows to 120GB and tried
>>>> again,
>>>> A minor miracle happened and mbr2gpt came back with  'Operation
>>>> successful'.
>>>>
>>>> Now that drive shows up as GPT at last, Next I ran Macrium reflect
>>>> 'Repair windows boot problems and that brought back all the missing
>>>> partitions on the disk.
>>>>
>>>> Whynot11 now shows all 11 blocks Green and Windows update shows
>>>> computer is compatible with Windows 11, So Sunday I will install 11,
>>>> Well, That that was a bit of a struggle, Thanks to all for your help.
>>>>
>>>> Rene
>>>>
>>>>
>>>>
>>>>
>>> Progess. Congratulations.
>>>   Revisit my post on the partition order for GPT.
>>>   - that order is the usual(and recommended) order.
>>>
>>> Looking back at your earlier error messages, in hindsight it may have
>>> helped to include or link to a picture of the partitions on the disk
>>> which showed their position and size.
>>>
>>> Even now it still might be appropriate to see a picture or the end
>>> result to determine if those GPT partitions are in the recommended
>>> order.
>>>   Why..in some cases after conversion the order isn't exactly as
>>> recommended. MSFT fixed a problem with the Recovery being created at
>>> the beginning of the disk and the EFI after the Windows which is
>>> contrary to the recommended GPT order(System-EFI, MSR, Windows,
>>> Recovery).
>>> - With the Recovery at the beginning, if the size of the Recovery is
>>> insufficient, it can not be resized larger thus an additional and the
>>> new current Recovery will be created in the recommended
>>> position(after the Windows) by shrinking the Windows partition -
>>> typically done if necessary with a feature update or o/s change(e.g.
>>> later 10 version/build and/or 10 to 11 o/s upgrade). Second, if a new
>>> EFI is created elswhere on the disk(after the Windows) then the end
>>> result could be a disk with a no longer user older System
>>> partiton(remnant of MBR at the beginning of the disk) and an extra an
>>> no longer used/necessary Recovery partition(at beginning or end of disk.
>>>
>>> With the partitions in the recommended order(System, MSR, Windows,
>>> Recovery - whenever the Recovery needs more space, by design Windows
>>> will shrink the Windows allowing the Recovery(adjacent and to right
>>> of Windows parition) to be resized larger preventing the recreation
>>> of a new Recovery partition.
>>>
>>> If you've time - Post or link to a picture or elaborate further on
>>> the partitions and order.
>>>
>>>
>> OK as i rememberit .can'yt accss till next week
>>
>> 1.  8GB unallocated
>> 2. 16GB other
>> 3. 100MB EFI
>
> 4. Windows C: 120GB
>
>> 5. 616MB recovery
>> 6. other data partitions
>>
>> I think that is fairly accurate, The 8GB came about when I moved
>> Windows over, so I think i will leave it alone,  I don't need the space.
>>
>> Rene
>
>
>> Screwed that up, Windows should have been 4.
>
> Rene
>
>
>
>
I thought the Windows partition might show up eventually :)
As intended, The EFI was created before Windows.
- also if one looks at that disks partitions in diskpart or a 3rd
party utility like Easus Partition Master or similar utility, the 16MB
MSR might exist between the EFI and Windows partitions.

And Recovery sits in the right place(after Windows).
- Interesting size too.

The oddballs are those at front
8 GB unallocated(understandable you moved Windows, though if that was
the case and Windows was first in the devices earlier days, with MBR the
System and Windows could have been the same partition.
The other odd ball is 16 GB of something.

Imo, if working then best left alone. Recapturing that 24 GB and moving
the EFI with 3rd party tools or intermingling with Diskpart and 3rd
party tools can get messy.
-lol..it might actually be easier to backup the data to external, wipe
to bear metal, diskpart for GPT(EFI, MSR, Windows, Recovery and data
partitions then clean install Windows to the Windows partition.

Paul may have ideas too.

--
....w¡ñ§±¤ñ

On 2/24/2022 12:38 AM, ...w¡ñ§±¤ñ wrote:
> Rene Lamontagne wrote:
>> On 2022-02-23 10:42 p.m., Rene Lamontagne wrote:
>>> On 2022-02-23 3:30 p.m., ...w¡ñ§±¤ñ wrote:
>>>> Rene Lamontagne wrote:
>>>>> On 2022-02-14 1:39 a.m., ...w¡ñ§±¤ñ wrote:
>>>>>> Rene Lamontagne wrote:
>>>>>>>
>>>>>>> Not enough room to create the EFI partition, I have about 65GB free on C: and 300GB of unallocated room on the drive so I am thinking this space needs to be before C:\
>>>>>>>
>>>>>>> My Son then needed the use of the machine so will continue next week and create another unallocated 8GB in front of C: maybe that will do it.
>>>>>>>
>>>>>>> Rene
>>>>>>>
>>>>>>>
>>>>>> GPT partiton order on disk
>>>>>> 1st EFI
>>>>>> 2nd MSR
>>>>>> 3rd Windows
>>>>>> 4th Recovery
>>>>>> 5th and later - anything else
>>>>>>
>>>>>> EFI 100 MB
>>>>>> MSR 16 MB
>>>>>> Windows - optional
>>>>>> Recovery 500 MB(1 GB is better)
>>>>>>
>>>>>>
>>>>>
>>>>> Yes some were missing so tried the mbr2gpt  thing again today, Now it says 'operation failed, could not shrink Windcows partion C,'
>>>>>
>>>>> Now why in hell did it want to do that?  C: is 100GB and there is over 350GB of unallocated space on the disk, tried again, same results
>>>>>
>>>>> Well to see what would happen I resized Windows to 120GB and tried again,
>>>>> A minor miracle happened and mbr2gpt came back with  'Operation successful'.
>>>>>
>>>>> Now that drive shows up as GPT at last, Next I ran Macrium reflect 'Repair windows boot problems and that brought back all the missing partitions on the disk.
>>>>>
>>>>> Whynot11 now shows all 11 blocks Green and Windows update shows computer is compatible with Windows 11, So Sunday I will install 11, Well, That that was a bit of a struggle, Thanks to all for your help.
>>>>>
>>>>> Rene
>>>>>
>>>>>
>>>>>
>>>>>
>>>> Progess. Congratulations.
>>>>   Revisit my post on the partition order for GPT.
>>>>   - that order is the usual(and recommended) order.
>>>>
>>>> Looking back at your earlier error messages, in hindsight it may have helped to include or link to a picture of the partitions on the disk which showed their position and size.
>>>>
>>>> Even now it still might be appropriate to see a picture or the end result to determine if those GPT partitions are in the recommended order.
>>>>   Why..in some cases after conversion the order isn't exactly as recommended. MSFT fixed a problem with the Recovery being created at the beginning of the disk and the EFI after the Windows which is contrary to the recommended GPT order(System-EFI, MSR, Windows, Recovery).
>>>> - With the Recovery at the beginning, if the size of the Recovery is insufficient, it can not be resized larger thus an additional and the new current Recovery will be created in the recommended position(after the Windows) by shrinking the Windows partition - typically done if necessary with a feature update or o/s change(e.g. later 10 version/build and/or 10 to 11 o/s upgrade). Second, if a new EFI is created elswhere on the disk(after the Windows) then the end result could be a disk with a no longer user older System partiton(remnant of MBR at the beginning of the disk) and an extra an no longer used/necessary Recovery partition(at beginning or end of disk.
>>>>
>>>> With the partitions in the recommended order(System, MSR, Windows, Recovery - whenever the Recovery needs more space, by design Windows will shrink the Windows allowing the Recovery(adjacent and to right of Windows parition) to be resized larger preventing the recreation of a new Recovery partition.
>>>>
>>>> If you've time - Post or link to a picture or elaborate further on the partitions and order.
>>>>
>>>>
>>> OK as i rememberit .can'yt accss till next week
>>>
>>> 1.  8GB unallocated
>>> 2. 16GB other
>>> 3. 100MB EFI
>>
>> 4. Windows C: 120GB
>>
>>> 5. 616MB recovery
>>> 6. other data partitions
>>>
>>> I think that is fairly accurate, The 8GB came about when I moved Windows over, so I think i will leave it alone,  I don't need the space.
>>>
>>> Rene
>>
>>
>>> Screwed that up, Windows should have been 4.
>>
>> Rene
>>
>>
>>
>>
> I thought the Windows partition might show up eventually :)
> As intended, The EFI was created before Windows.
>  - also if one looks at that disks partitions in diskpart or a 3rd party utility like Easus Partition Master or similar utility, the 16MB MSR might exist between the EFI and Windows partitions.
>
> And Recovery sits in the right place(after Windows).
>  - Interesting size too.
>
> The oddballs are those at front
>  8 GB unallocated(understandable you moved Windows, though if that was the case and Windows was first in the devices earlier days, with MBR the System and Windows could have been the same partition.
> The other odd ball is 16 GB of something.
>
> Imo, if working then best left alone. Recapturing that 24 GB and moving the EFI with 3rd party tools or intermingling with Diskpart and 3rd party tools can get messy.
>  -lol..it might actually be easier to backup the data to external, wipe to bear metal, diskpart for GPT(EFI, MSR, Windows, Recovery and data partitions then clean install Windows to the Windows partition.
>
> Paul may have ideas too.

Depending on what is in those "other" partitions, they may be
getting auto-assigned drive letters, or they may receive fixed
letters by the user.

If for any reason, some software that happened to run in there, was
addressed by letter, you could break something by moving them.

I think I was screwing around here recently (with several backups available on
the 6TB scratch), and I noticed a Windows partition change from
C: to D: and I took that as a bad omen (the OS was not running
at the time). The hardware discovery process plus the auto-assignment
of letters, occasionally leads to surprises, and you have to be careful
to not move too much stuff around.

*******

As for the 16MB partition, that little pecker is a menace.

The MSR or Microsoft System Reserved, is merely a buffer space.
It has no valid partition inside.

If I use a tool like Paragon on it, as soon as no recognized partition
is located inside, it's treated as if it is "corrupted". Lots of
tools run CHKDSK as part of their routine, and that cannot end well.
I think Paragon and Gparted, gave me a hard time. My operation would
not complete.

To move the MSR, you can use Macrium. Back up the MSR to a .mrimg,
then drag and drop into the correct "space" in your partition lineup.

I have been redoing ("cleaning") a 3TB drive here by hand, and that's how I
handled the MSR. Drag and drop restore. The only reason the MSR
needed to be moved, is there was some space to consolidate by
moving C: to the left. And the MSR had to be moved left-ward and
stop acting as a "barrier" to my partition management techniques.
I went from 24 partitions, to 7 partitions.

The MSR does not show up in all tools. And that's part of the fun.

In Linux, you can use

sudo gdisk /dev/sdb
p
q

and print out the partitions on your GPT disk. For a legacy MSDOS
partitioned disk, you can try this (when Gdisk patiently explains to
you that it is not a GPT disk).

sudo fdisk /dev/sdb
p
q

Do not start typing too quick in Gdisk, as it can throw up a prompt
after you start it, and the prompt needs to be answered. Most of the
time there is no prompt, and you could be in a rush to type p for "print".
Fdisk leaves fewer manhole covers off the holes :-) It's a more reliable
friend.

The gdisk is a great design, and a help. I don't have a problem with the
service it provides. I also don't see a good way of handling the prompt issue.
I don't know how I'd fix that, if I was editing the source.

Paul

On 2022-02-24 12:07 a.m., Paul wrote:
> On 2/24/2022 12:38 AM, ...w¡ñ§±¤ñ wrote:
>> Rene Lamontagne wrote:
>>> On 2022-02-23 10:42 p.m., Rene Lamontagne wrote:
>>>> On 2022-02-23 3:30 p.m., ...w¡ñ§±¤ñ wrote:
>>>>> Rene Lamontagne wrote:
>>>>>> On 2022-02-14 1:39 a.m., ...w¡ñ§±¤ñ wrote:
>>>>>>> Rene Lamontagne wrote:
>>>>>>>>
>>>>>>>> Not enough room to create the EFI partition, I have about 65GB
>>>>>>>> free on C: and 300GB of unallocated room on the drive so I am
>>>>>>>> thinking this space needs to be before C:\
>>>>>>>>
>>>>>>>> My Son then needed the use of the machine so will continue next
>>>>>>>> week and create another unallocated 8GB in front of C: maybe
>>>>>>>> that will do it.
>>>>>>>>
>>>>>>>> Rene
>>>>>>>>
>>>>>>>>
>>>>>>> GPT partiton order on disk
>>>>>>> 1st EFI
>>>>>>> 2nd MSR
>>>>>>> 3rd Windows
>>>>>>> 4th Recovery
>>>>>>> 5th and later - anything else
>>>>>>>
>>>>>>> EFI 100 MB
>>>>>>> MSR 16 MB
>>>>>>> Windows - optional
>>>>>>> Recovery 500 MB(1 GB is better)
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> Yes some were missing so tried the mbr2gpt  thing again today, Now
>>>>>> it says 'operation failed, could not shrink Windcows partion C,'
>>>>>>
>>>>>> Now why in hell did it want to do that?  C: is 100GB and there is
>>>>>> over 350GB of unallocated space on the disk, tried again, same
>>>>>> results
>>>>>>
>>>>>> Well to see what would happen I resized Windows to 120GB and tried
>>>>>> again,
>>>>>> A minor miracle happened and mbr2gpt came back with  'Operation
>>>>>> successful'.
>>>>>>
>>>>>> Now that drive shows up as GPT at last, Next I ran Macrium reflect
>>>>>> 'Repair windows boot problems and that brought back all the
>>>>>> missing partitions on the disk.
>>>>>>
>>>>>> Whynot11 now shows all 11 blocks Green and Windows update shows
>>>>>> computer is compatible with Windows 11, So Sunday I will install
>>>>>> 11, Well, That that was a bit of a struggle, Thanks to all for
>>>>>> your help.
>>>>>>
>>>>>> Rene
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>> Progess. Congratulations.
>>>>>   Revisit my post on the partition order for GPT.
>>>>>   - that order is the usual(and recommended) order.
>>>>>
>>>>> Looking back at your earlier error messages, in hindsight it may
>>>>> have helped to include or link to a picture of the partitions on
>>>>> the disk which showed their position and size.
>>>>>
>>>>> Even now it still might be appropriate to see a picture or the end
>>>>> result to determine if those GPT partitions are in the recommended
>>>>> order.
>>>>>   Why..in some cases after conversion the order isn't exactly as
>>>>> recommended. MSFT fixed a problem with the Recovery being created
>>>>> at the beginning of the disk and the EFI after the Windows which is
>>>>> contrary to the recommended GPT order(System-EFI, MSR, Windows,
>>>>> Recovery).
>>>>> - With the Recovery at the beginning, if the size of the Recovery
>>>>> is insufficient, it can not be resized larger thus an additional
>>>>> and the new current Recovery will be created in the recommended
>>>>> position(after the Windows) by shrinking the Windows partition -
>>>>> typically done if necessary with a feature update or o/s
>>>>> change(e.g. later 10 version/build and/or 10 to 11 o/s upgrade).
>>>>> Second, if a new EFI is created elswhere on the disk(after the
>>>>> Windows) then the end result could be a disk with a no longer user
>>>>> older System partiton(remnant of MBR at the beginning of the disk)
>>>>> and an extra an no longer used/necessary Recovery partition(at
>>>>> beginning or end of disk.
>>>>>
>>>>> With the partitions in the recommended order(System, MSR, Windows,
>>>>> Recovery - whenever the Recovery needs more space, by design
>>>>> Windows will shrink the Windows allowing the Recovery(adjacent and
>>>>> to right of Windows parition) to be resized larger preventing the
>>>>> recreation of a new Recovery partition.
>>>>>
>>>>> If you've time - Post or link to a picture or elaborate further on
>>>>> the partitions and order.
>>>>>
>>>>>
>>>> OK as i rememberit .can'yt accss till next week
>>>>
>>>> 1.  8GB unallocated
>>>> 2. 16GB other
>>>> 3. 100MB EFI
>>>
>>> 4. Windows C: 120GB
>>>
>>>> 5. 616MB recovery
>>>> 6. other data partitions
>>>>
>>>> I think that is fairly accurate, The 8GB came about when I moved
>>>> Windows over, so I think i will leave it alone,  I don't need the
>>>> space.
>>>>
>>>> Rene
>>>
>>>
>>>> Screwed that up, Windows should have been 4.
>>>
>>> Rene
>>>
>>>
>>>
>>>
>> I thought the Windows partition might show up eventually :)
>> As intended, The EFI was created before Windows.
>>   - also if one looks at that disks partitions in diskpart or a 3rd
>> party utility like Easus Partition Master or similar utility, the 16MB
>> MSR might exist between the EFI and Windows partitions.
>>
>> And Recovery sits in the right place(after Windows).
>>   - Interesting size too.
>>
>> The oddballs are those at front
>>   8 GB unallocated(understandable you moved Windows, though if that
>> was the case and Windows was first in the devices earlier days, with
>> MBR the System and Windows could have been the same partition.
>> The other odd ball is 16 GB of something.
>>
>> Imo, if working then best left alone. Recapturing that 24 GB and
>> moving the EFI with 3rd party tools or intermingling with Diskpart and
>> 3rd party tools can get messy.
>>   -lol..it might actually be easier to backup the data to external,
>> wipe to bear metal, diskpart for GPT(EFI, MSR, Windows, Recovery and
>> data partitions then clean install Windows to the Windows partition.
>>
>> Paul may have ideas too.
>
> Depending on what is in those "other" partitions, they may be
> getting auto-assigned drive letters, or they may receive fixed
> letters by the user.
>
> If for any reason, some software that happened to run in there, was
> addressed by letter, you could break something by moving them.
>
> I think I was screwing around here recently (with several backups
> available on
> the 6TB scratch), and I noticed a Windows partition change from
> C: to D: and I took that as a bad omen (the OS was not running
> at the time). The hardware discovery process plus the auto-assignment
> of letters, occasionally leads to surprises, and you have to be careful
> to not move too much stuff around.
>
> *******
>
> As for the 16MB partition, that little pecker is a menace.
>
> The MSR or Microsoft System Reserved, is merely a buffer space.
> It has no valid partition inside.
>
> If I use a tool like Paragon on it, as soon as no recognized partition
> is located inside, it's treated as if it is "corrupted". Lots of
> tools run CHKDSK as part of their routine, and that cannot end well.
> I think Paragon and Gparted, gave me a hard time. My operation would
> not complete.
>
> To move the MSR, you can use Macrium. Back up the MSR to a .mrimg,
> then drag and drop into the correct "space" in your partition lineup.
>
> I have been redoing ("cleaning") a 3TB drive here by hand, and that's how I
> handled the MSR. Drag and drop restore. The only reason the MSR
> needed to be moved, is there was some space to consolidate by
> moving C: to the left. And the MSR had to be moved left-ward and
> stop acting as a "barrier" to my partition management techniques.
> I went from 24 partitions, to 7 partitions.
>
> The MSR does not show up in all tools. And that's part of the fun.
>
> In Linux, you can use
>
>    sudo gdisk /dev/sdb
>    p
>    q
>
> and print out the partitions on your GPT disk. For a legacy MSDOS
> partitioned disk, you can try this (when Gdisk patiently explains to
> you that it is not a GPT disk).
>
>    sudo fdisk /dev/sdb
>    p
>    q
>
> Do not start typing too quick in Gdisk, as it can throw up a prompt
> after you start it, and the prompt needs to be answered. Most of the
> time there is no prompt, and you could be in a rush to type p for "print".
> Fdisk leaves fewer manhole covers off the holes :-) It's a more reliable
> friend.
>
> The gdisk is a great design, and a help. I don't have a problem with the
> service it provides. I also don't see a good way of handling the prompt
> issue.
> I don't know how I'd fix that, if I was editing the source.
>
>    Paul

On 2022-02-24 3:22 p.m., Rene Lamontagne wrote:
> On 2022-02-24 12:07 a.m., Paul wrote:
>> On 2/24/2022 12:38 AM, ...w¡ñ§±¤ñ wrote:
>>> Rene Lamontagne wrote:
>>>> On 2022-02-23 10:42 p.m., Rene Lamontagne wrote:
>>>>> On 2022-02-23 3:30 p.m., ...w¡ñ§±¤ñ wrote:
>>>>>> Rene Lamontagne wrote:
>>>>>>> On 2022-02-14 1:39 a.m., ...w¡ñ§±¤ñ wrote:
>>>>>>>> Rene Lamontagne wrote:
>>>>>>>>>
>>>>>>>>> Not enough room to create the EFI partition, I have about 65GB
>>>>>>>>> free on C: and 300GB of unallocated room on the drive so I am
>>>>>>>>> thinking this space needs to be before C:\
>>>>>>>>>
>>>>>>>>> My Son then needed the use of the machine so will continue next
>>>>>>>>> week and create another unallocated 8GB in front of C: maybe
>>>>>>>>> that will do it.
>>>>>>>>>
>>>>>>>>> Rene
>>>>>>>>>
>>>>>>>>>
>>>>>>>> GPT partiton order on disk
>>>>>>>> 1st EFI
>>>>>>>> 2nd MSR
>>>>>>>> 3rd Windows
>>>>>>>> 4th Recovery
>>>>>>>> 5th and later - anything else
>>>>>>>>
>>>>>>>> EFI 100 MB
>>>>>>>> MSR 16 MB
>>>>>>>> Windows - optional
>>>>>>>> Recovery 500 MB(1 GB is better)
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>> Yes some were missing so tried the mbr2gpt  thing again today,
>>>>>>> Now it says 'operation failed, could not shrink Windcows partion C,'
>>>>>>>
>>>>>>> Now why in hell did it want to do that?  C: is 100GB and there is
>>>>>>> over 350GB of unallocated space on the disk, tried again, same
>>>>>>> results
>>>>>>>
>>>>>>> Well to see what would happen I resized Windows to 120GB and
>>>>>>> tried again,
>>>>>>> A minor miracle happened and mbr2gpt came back with  'Operation
>>>>>>> successful'.
>>>>>>>
>>>>>>> Now that drive shows up as GPT at last, Next I ran Macrium
>>>>>>> reflect 'Repair windows boot problems and that brought back all
>>>>>>> the missing partitions on the disk.
>>>>>>>
>>>>>>> Whynot11 now shows all 11 blocks Green and Windows update shows
>>>>>>> computer is compatible with Windows 11, So Sunday I will install
>>>>>>> 11, Well, That that was a bit of a struggle, Thanks to all for
>>>>>>> your help.
>>>>>>>
>>>>>>> Rene
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>> Progess. Congratulations.
>>>>>>   Revisit my post on the partition order for GPT.
>>>>>>   - that order is the usual(and recommended) order.
>>>>>>
>>>>>> Looking back at your earlier error messages, in hindsight it may
>>>>>> have helped to include or link to a picture of the partitions on
>>>>>> the disk which showed their position and size.
>>>>>>
>>>>>> Even now it still might be appropriate to see a picture or the end
>>>>>> result to determine if those GPT partitions are in the recommended
>>>>>> order.
>>>>>>   Why..in some cases after conversion the order isn't exactly as
>>>>>> recommended. MSFT fixed a problem with the Recovery being created
>>>>>> at the beginning of the disk and the EFI after the Windows which
>>>>>> is contrary to the recommended GPT order(System-EFI, MSR, Windows,
>>>>>> Recovery).
>>>>>> - With the Recovery at the beginning, if the size of the Recovery
>>>>>> is insufficient, it can not be resized larger thus an additional
>>>>>> and the new current Recovery will be created in the recommended
>>>>>> position(after the Windows) by shrinking the Windows partition -
>>>>>> typically done if necessary with a feature update or o/s
>>>>>> change(e.g. later 10 version/build and/or 10 to 11 o/s upgrade).
>>>>>> Second, if a new EFI is created elswhere on the disk(after the
>>>>>> Windows) then the end result could be a disk with a no longer user
>>>>>> older System partiton(remnant of MBR at the beginning of the disk)
>>>>>> and an extra an no longer used/necessary Recovery partition(at
>>>>>> beginning or end of disk.
>>>>>>
>>>>>> With the partitions in the recommended order(System, MSR, Windows,
>>>>>> Recovery - whenever the Recovery needs more space, by design
>>>>>> Windows will shrink the Windows allowing the Recovery(adjacent and
>>>>>> to right of Windows parition) to be resized larger preventing the
>>>>>> recreation of a new Recovery partition.
>>>>>>
>>>>>> If you've time - Post or link to a picture or elaborate further on
>>>>>> the partitions and order.
>>>>>>
>>>>>>
>>>>> OK as i rememberit .can'yt accss till next week
>>>>>
>>>>> 1.  8GB unallocated
>>>>> 2. 16GB other
>>>>> 3. 100MB EFI
>>>>
>>>> 4. Windows C: 120GB
>>>>
>>>>> 5. 616MB recovery
>>>>> 6. other data partitions
>>>>>
>>>>> I think that is fairly accurate, The 8GB came about when I moved
>>>>> Windows over, so I think i will leave it alone,  I don't need the
>>>>> space.
>>>>>
>>>>> Rene
>>>>
>>>>
>>>>> Screwed that up, Windows should have been 4.
>>>>
>>>> Rene
>>>>
>>>>
>>>>
>>>>
>>> I thought the Windows partition might show up eventually :)
>>> As intended, The EFI was created before Windows.
>>>   - also if one looks at that disks partitions in diskpart or a 3rd
>>> party utility like Easus Partition Master or similar utility, the
>>> 16MB MSR might exist between the EFI and Windows partitions.
>>>
>>> And Recovery sits in the right place(after Windows).
>>>   - Interesting size too.
>>>
>>> The oddballs are those at front
>>>   8 GB unallocated(understandable you moved Windows, though if that
>>> was the case and Windows was first in the devices earlier days, with
>>> MBR the System and Windows could have been the same partition.
>>> The other odd ball is 16 GB of something.
>>>
>>> Imo, if working then best left alone. Recapturing that 24 GB and
>>> moving the EFI with 3rd party tools or intermingling with Diskpart
>>> and 3rd party tools can get messy.
>>>   -lol..it might actually be easier to backup the data to external,
>>> wipe to bear metal, diskpart for GPT(EFI, MSR, Windows, Recovery and
>>> data partitions then clean install Windows to the Windows partition.
>>>
>>> Paul may have ideas too.
>>
>> Depending on what is in those "other" partitions, they may be
>> getting auto-assigned drive letters, or they may receive fixed
>> letters by the user.
>>
>> If for any reason, some software that happened to run in there, was
>> addressed by letter, you could break something by moving them.
>>
>> I think I was screwing around here recently (with several backups
>> available on
>> the 6TB scratch), and I noticed a Windows partition change from
>> C: to D: and I took that as a bad omen (the OS was not running
>> at the time). The hardware discovery process plus the auto-assignment
>> of letters, occasionally leads to surprises, and you have to be careful
>> to not move too much stuff around.
>>
>> *******
>>
>> As for the 16MB partition, that little pecker is a menace.
>>
>> The MSR or Microsoft System Reserved, is merely a buffer space.
>> It has no valid partition inside.
>>
>> If I use a tool like Paragon on it, as soon as no recognized partition
>> is located inside, it's treated as if it is "corrupted". Lots of
>> tools run CHKDSK as part of their routine, and that cannot end well.
>> I think Paragon and Gparted, gave me a hard time. My operation would
>> not complete.
>>
>> To move the MSR, you can use Macrium. Back up the MSR to a .mrimg,
>> then drag and drop into the correct "space" in your partition lineup.
>>
>> I have been redoing ("cleaning") a 3TB drive here by hand, and that's
>> how I
>> handled the MSR. Drag and drop restore. The only reason the MSR
>> needed to be moved, is there was some space to consolidate by
>> moving C: to the left. And the MSR had to be moved left-ward and
>> stop acting as a "barrier" to my partition management techniques.
>> I went from 24 partitions, to 7 partitions.
>>
>> The MSR does not show up in all tools. And that's part of the fun.
>>
>> In Linux, you can use
>>
>>     sudo gdisk /dev/sdb
>>     p
>>     q
>>
>> and print out the partitions on your GPT disk. For a legacy MSDOS
>> partitioned disk, you can try this (when Gdisk patiently explains to
>> you that it is not a GPT disk).
>>
>>     sudo fdisk /dev/sdb
>>     p
>>     q
>>
>> Do not start typing too quick in Gdisk, as it can throw up a prompt
>> after you start it, and the prompt needs to be answered. Most of the
>> time there is no prompt, and you could be in a rush to type p for
>> "print".
>> Fdisk leaves fewer manhole covers off the holes :-) It's a more reliable
>> friend.
>>
>> The gdisk is a great design, and a help. I don't have a problem with the
>> service it provides. I also don't see a good way of handling the
>> prompt issue.
>> I don't know how I'd fix that, if I was editing the source.
>>
>>     Paul
>
> Thanks Winston and Paul for all the help and advice.
>
> If it was my own machine I would probably try moving stuff around at my
> leisure, But seeing it is my Sons and I have only limited time on it and
> he depends quite heavily on its use, I think I will leave it be as it is
> now working quite well,
> If it were to act up badly again I would then do the nuke and pave game
> with a fresh start.
>
> Rene
>

On 2/27/2022 4:29 PM, Rene Lamontagne wrote:

>
> Last Lap, first thing, Macrium image backup of Windows 10, Then to update where message was do you want to install Windows 11 on this machine?  YES.
>
> One hour later it was all done, everything seems Fine, Then another Macrium image back up of Windows 11 system.
>
> All Done, End of Story.  :-)
>
> Rene

I guess we'll find out later, whether this W11 idea
was a good one or not.

I'm not liking the Notepad change, because the cursor
disappears after maybe ten seconds. Now, who thought
that would be a good idea ? What school of software
design teaches that ?

Paul

On 2022-02-28 00:00, Paul wrote:
> On 2/27/2022 4:29 PM, Rene Lamontagne wrote:

> I'm not liking the Notepad change, because the cursor
> disappears after maybe ten seconds. Now, who thought
> that would be a good idea ? What school of software
> design teaches that ?

In small and controlled machines, blinking a cursor means having the
display active, calculating and doing that tiny change continuously.
Now, on a pure text display this can be done by the display hardware.
What about a graphical display? Maybe not, because the cursor depends on
the font, and in the worse case it involves the CPU. Maybe, maybe not.

So it can be a small optimization to conserve power, that would make
sense if there are another myriad of small changes with the same aim.
For instance, it does not make sense to stop the cursor blink, if on
another part of the display the clock is updating to the second or there
is a moving graph.

--
Cheers, Carlos.

On 2/27/2022 6:42 PM, Carlos E.R. wrote:
> On 2022-02-28 00:00, Paul wrote:
>> On 2/27/2022 4:29 PM, Rene Lamontagne wrote:
>
>
>> I'm not liking the Notepad change, because the cursor
>> disappears after maybe ten seconds. Now, who thought
>> that would be a good idea ? What school of software
>> design teaches that ?
>
> In small and controlled machines, blinking a cursor means having the display active, calculating and doing that tiny change continuously. Now, on a pure text display this can be done by the display hardware. What about a graphical display? Maybe not, because the cursor depends on the font, and in the worse case it involves the CPU. Maybe, maybe not.
>
> So it can be a small optimization to conserve power, that would make sense if there are another myriad of small changes with the same aim. For instance, it does not make sense to stop the cursor blink, if on another part of the display the clock is updating to the second or there is a moving graph.

Today, we don't really know what acceleration functions
are done by dedicated hardware.

Some of what the mouse cursor does, is supported in hardware,
and is an overlay in front of the rest of the composited desktop.

For an inserted i-beam, I don't know if there is a separate
hardware feature for that or not. 2D acceleration and BITBLT
development stopped decades ago, which is why it will be hard
to trace whether any new features have been added on the sly.

Normally features like this get added, because "it's such a
nuisance to keep having to code up these features".

There was a time, before NVidia, when there were "sprites".
Which is an example of an animation technology with position
and velocity and also animation of the pixmap. But whether there
are any sprites in regular video cards today, not a clue. Texas
Instruments made a sprite chip a long time ago, and a manager
at work gave us a demo of his sprite project. The colors
were awful. The CPU could go to sleep, while a sprite
moved on a diagonal across the screen. Obviously intended
for things like "Asteroids".

Paul

On 2022-02-28 01:39, Paul wrote:
> On 2/27/2022 6:42 PM, Carlos E.R. wrote:
>> On 2022-02-28 00:00, Paul wrote:
>>> On 2/27/2022 4:29 PM, Rene Lamontagne wrote:
>>
>>
>>> I'm not liking the Notepad change, because the cursor
>>> disappears after maybe ten seconds. Now, who thought
>>> that would be a good idea ? What school of software
>>> design teaches that ?
>>
>> In small and controlled machines, blinking a cursor means having the
>> display active, calculating and doing that tiny change continuously.
>> Now, on a pure text display this can be done by the display hardware.
>> What about a graphical display? Maybe not, because the cursor depends
>> on the font, and in the worse case it involves the CPU. Maybe, maybe not.
>>
>> So it can be a small optimization to conserve power, that would make
>> sense if there are another myriad of small changes with the same aim.
>> For instance, it does not make sense to stop the cursor blink, if on
>> another part of the display the clock is updating to the second or
>> there is a moving graph.
>
> Today, we don't really know what acceleration functions
> are done by dedicated hardware.
>
> Some of what the mouse cursor does, is supported in hardware,
> and is an overlay in front of the rest of the composited desktop.
>
> For an inserted i-beam, I don't know if there is a separate
> hardware feature for that or not. 2D acceleration and BITBLT
> development stopped decades ago, which is why it will be hard
> to trace whether any new features have been added on the sly.
>
> Normally features like this get added, because "it's such a
> nuisance to keep having to code up these features".
>
> There was a time, before NVidia, when there were "sprites".
> Which is an example of an animation technology with position
> and velocity and also animation of the pixmap. But whether there
> are any sprites in regular video cards today, not a clue. Texas
> Instruments made a sprite chip a long time ago, and a manager
> at work gave us a demo of his sprite project. The colors
> were awful. The CPU could go to sleep, while a sprite
> moved on a diagonal across the screen. Obviously intended
> for things like "Asteroids".

Possibly what we call "hardware" in a graphics card is a dedicated GPU
with memory, that is, a CPU specialized in graphical operations.

Dunno, maybe someone has measured how many miliwats takes to blink the
cursor :-)

--
Cheers, Carlos.

On 2/27/2022 10:24 PM, Carlos E.R. wrote:
> On 2022-02-28 01:39, Paul wrote:
>> On 2/27/2022 6:42 PM, Carlos E.R. wrote:
>>> On 2022-02-28 00:00, Paul wrote:
>>>> On 2/27/2022 4:29 PM, Rene Lamontagne wrote:
>>>
>>>
>>>> I'm not liking the Notepad change, because the cursor
>>>> disappears after maybe ten seconds. Now, who thought
>>>> that would be a good idea ? What school of software
>>>> design teaches that ?
>>>
>>> In small and controlled machines, blinking a cursor means having the display active, calculating and doing that tiny change continuously. Now, on a pure text display this can be done by the display hardware. What about a graphical display? Maybe not, because the cursor depends on the font, and in the worse case it involves the CPU. Maybe, maybe not.
>>>
>>> So it can be a small optimization to conserve power, that would make sense if there are another myriad of small changes with the same aim. For instance, it does not make sense to stop the cursor blink, if on another part of the display the clock is updating to the second or there is a moving graph.
>>
>> Today, we don't really know what acceleration functions
>> are done by dedicated hardware.
>>
>> Some of what the mouse cursor does, is supported in hardware,
>> and is an overlay in front of the rest of the composited desktop.
>>
>> For an inserted i-beam, I don't know if there is a separate
>> hardware feature for that or not. 2D acceleration and BITBLT
>> development stopped decades ago, which is why it will be hard
>> to trace whether any new features have been added on the sly.
>>
>> Normally features like this get added, because "it's such a
>> nuisance to keep having to code up these features".
>>
>> There was a time, before NVidia, when there were "sprites".
>> Which is an example of an animation technology with position
>> and velocity and also animation of the pixmap. But whether there
>> are any sprites in regular video cards today, not a clue. Texas
>> Instruments made a sprite chip a long time ago, and a manager
>> at work gave us a demo of his sprite project. The colors
>> were awful. The CPU could go to sleep, while a sprite
>> moved on a diagonal across the screen. Obviously intended
>> for things like "Asteroids".
>
> Possibly what we call "hardware" in a graphics card is a
> dedicated GPU with memory, that is, a CPU specialized in graphical operations.
>
> Dunno, maybe someone has measured how many miliwats takes to blink the cursor :-)

Unless it causes the GPU to switch from the idling clock (100MHz)
to the 3D clock (1500MHz), the incremental power required should be unmeasurable.

Circuits with 10 billion gate equivalents have "leakage currents",
and those would be more significant than any cursor blink. That's
what made the Prescott CPU such a menace - 25% of the power was
wasted doing absolutely nothing. It just leaked through the
totem pole structure.

This is why modern CMOS has a mixture of two transistor and three
transistor output stages. The three transistor ones can gate off
leakage when they're idle. The conventional two transistor ones
are for speed.

We reached a "golden peak" maybe ten years ago, where an idle
GPU used only 3W of power, and that might well have been used
to keep the VRAM refreshed. More modern cards no longer
sip power like that, and I would guess part of the blame
is 5nm or 7nm feature sizes. Maybe now the idle has
sneaked up to 13W or 20W, at 100MHz. The GPU inside your
CPU, being a smaller die, won't be quite that bad.

There is a plan for even-more-power-consuming GPUs,
requiring a new PCIe power connector. Without looking it
up, I seem to remember 350W being suggested as coming soon.

Paul

On 2022-02-28 04:43, Paul wrote:
> On 2/27/2022 10:24 PM, Carlos E.R. wrote:
>> On 2022-02-28 01:39, Paul wrote:
>>> On 2/27/2022 6:42 PM, Carlos E.R. wrote:
>>>> On 2022-02-28 00:00, Paul wrote:
>>>>> On 2/27/2022 4:29 PM, Rene Lamontagne wrote:
>>>>
>>>>
>>>>> I'm not liking the Notepad change, because the cursor
>>>>> disappears after maybe ten seconds. Now, who thought
>>>>> that would be a good idea ? What school of software
>>>>> design teaches that ?

....

>>> There was a time, before NVidia, when there were "sprites".
>>> Which is an example of an animation technology with position
>>> and velocity and also animation of the pixmap. But whether there
>>> are any sprites in regular video cards today, not a clue. Texas
>>> Instruments made a sprite chip a long time ago, and a manager
>>> at work gave us a demo of his sprite project. The colors
>>> were awful. The CPU could go to sleep, while a sprite
>>> moved on a diagonal across the screen. Obviously intended
>>> for things like "Asteroids".
>>
>> Possibly what we call "hardware" in a graphics card is a
>> dedicated GPU with memory, that is, a CPU specialized in graphical
>> operations.
>>
>> Dunno, maybe someone has measured how many miliwats takes to blink the
>> cursor :-)
>
> Unless it causes the GPU to switch from the idling clock (100MHz)
> to the 3D clock (1500MHz), the incremental power required should be
> unmeasurable.

It may become interesting in a laptop with hybrid hardware like Nvidia
Optimus.

>
> Circuits with 10 billion gate equivalents have "leakage currents",
> and those would be more significant than any cursor blink. That's
> what made the Prescott CPU such a menace - 25% of the power was
> wasted doing absolutely nothing. It just leaked through the
> totem pole structure.

Yep.

>
> This is why modern CMOS has a mixture of two transistor and three
> transistor output stages. The three transistor ones can gate off
> leakage when they're idle. The conventional two transistor ones
> are for speed.
>
> We reached a "golden peak" maybe ten years ago, where an idle
> GPU used only 3W of power, and that might well have been used
> to keep the VRAM refreshed. More modern cards no longer
> sip power like that, and I would guess part of the blame
> is 5nm or 7nm feature sizes. Maybe now the idle has
> sneaked up to 13W or 20W, at 100MHz. The GPU inside your
> CPU, being a smaller die, won't be quite that bad.
>
> There is a plan for even-more-power-consuming GPUs,
> requiring a new PCIe power connector. Without looking it
> up, I seem to remember 350W being suggested as coming soon.

Yagh.

--
Cheers, Carlos.

Rene Lamontagne wrote:
>
> Last Lap, first thing, Macrium image backup of Windows 10, Then to
> update where message was do you want to install Windows 11 on this
> machine?  YES.
>
> One hour later it was all done, everything seems Fine, Then another
> Macrium image back up of Windows 11 system.
>
> All Done, End of Story.  :-)
>
> Rene
>
Thanks for the update.

An hour to upgrade from some capable systems from W10 to W11 doesn't
seem that unusual.

On my Acer 8th Gen i7, the upgrade took a little less than 40 minutes
via WU.
The same device, ensuring digital licensing was intact and post Win11
was reverted via an Acronis 2020 image back to Win10 Pro and upgraded
via USB 3.1c Win11 media created with the Media Creation Tool via
setup.exe from within Win10 Pro.
- approx 70 minutes.

Wiping the same device, and discounting the time I booted the W11 media
and used a diskpart script to setup the four GPT required partitions(100
MB EFI, 16 MB MSR, Windows(all space not used for the other three
partitions), and a 1 GB Recovery) on a 1 TB SSD prior to clean
installing W11 was about 50 minutes.

--
....w¡ñ§±¤ñ

Paul <nospam@needed.invalid> wrote:
> On 2/27/2022 4:29 PM, Rene Lamontagne wrote:
>
>>
>> Last Lap, first thing, Macrium image backup of Windows 10, Then to
>> update where message was do you want to install Windows 11 on this machine?  YES.
>>
>> One hour later it was all done, everything seems Fine, Then another
>> Macrium image back up of Windows 11 system.
>>
>> All Done, End of Story.  :-)
>>
>> Rene
>
> I guess we'll find out later, whether this W11 idea
> was a good one or not.
>
> I'm not liking the Notepad change, because the cursor
> disappears after maybe ten seconds. Now, who thought
> that would be a good idea ? What school of software
> design teaches that ?
>
> Paul
>
>
>
>

Maybe Notepad++ is better

--
Zaidy036