Solved Weird HDD space issue…

[0pTicaL]

Weird HDD space issue…

I have two 1TB storage drives in a Lian Li EX-503 setup as individual disks. One is a old WD 1TB Black and one is a newer (2012) Hitachi. These drives are not in RAID1 but I have them both synced with FreeFileSync.

The weird thing is the WD has 1.05GB less space than the Hitachi and both drives have exactly the same amount of data (950,506,779,339 bytes). I do not have system restore turned on and I checked for hidden files and folders. There are no partitions on either drives.

*edit* - TreeSize also shows the exact same amount of space used on both drives.

I can't figure out where the 1.05GB on the WD drive went.

The only thing I can think of is the Hitachi uses Advanced Format and the WD does not, does AF affect available storage space?

 

[TVeblen]

A lot of it probably has to do with they way individual programs calculate disk size. What program/utility are you using to see this difference?

A possibility is that one drive was initialized and formatted by Windows, and the other by a Linux based program like Partition Wizard.
Windows will create blank space on either side of the volume for it's use in disk organization (Dynamic Disks and such). Linux will use the entire disk.

 

[0pTicaL]

A lot of it probably has to do with they way individual programs calculate disk size. What program/utility are you using to see this difference?

Not using any program. Just using Windows Properties (right-click > Properties) to see Used, Free and Capacity.

A possibility is that one drive was initialized and formatted by Windows, and the other by a Linux based program like Partition Wizard.
Windows will create blank space on either side of the volume for it's use in disk organization (Dynamic Disks and such). Linux will use the entire disk.

Nope, both drives were formatted under Windows 7 x64.

There's nothing complicated about this issue other than one drive not showing the correct amount of free space.

It's like this I have two 1TB HDDs inside a Lian Li EX-503 enclosure, the drives are setup as individual disks.

1) WD 1TB has 3 folders with files in them, the total size of all 3 folders 950,506,779,339 bytes

2) Hitachi 1TB is identical to the WD with 950,506,779,339 bytes

However the WD is showing 45.3GB of free space while the Hitachi is showing 46.3GB free space.

There are no hidden files or folders on either drive (other than Recycling Bin and System Volume Information). System Restore is disabled.

How can both drives use the same amount of space but have different amount of free space remaining?

 

[Britton30]

What does total capacity show for each drive? One may have 1 trillion bytes while the other may be a little shy but they market it as a 1TB drive.

 

[0pTicaL]

Both drives show a total capacity of 1,000,202,039,296 bytes (931GiB).

I have never seen a 1TB consumer drive, after format, have less or more than 1,000,202,039,296 bytes.

 

[Britton30]

Well it is odd then, the total capacity is the same whether raw, formatted, or full of data.
Could one drive have a hidden Folder such as a $RECYCLEBIN?
Here's why I asked, these are two Seagate 500GB Barracuda drives and capacity is different by 6MB and a bit.

 

[0pTicaL]

Here's a screenshot of the 1TBs. As you can see I resolved the issue, both show 45.3GB free space now. There seemed to be a anomaly with the Hitachi drive because the total size of all 3 folders exceeded the total used space shown under Drive Properties.

Both drives have the hidden folders $RECYCLE.BIN and System Volume Information but they are do not contain any data in them as shown under TreeSize. TreeSize is required to view the contents of System Volume Information because the folder is system locked.

I went ahead and deleted the Hitachi volume under Disk Management and quick formatted the drive and re-transferred the data from the WD to the Hitachi drive and now it shows up correctly.

However, there is still a 40.21MB difference.

 

[Britton30]

I don't know the answer of the used space mismatch. It may have something to do with how Windows measures sizes. Here is a snip of a folder I keep, well see the title, things in. The Size and Size on Disk are different as are all files.
Are both drivse formatted with the default sector size of 4096K?

 

[0pTicaL]

I don't know the answer of the used space mismatch. It may have something to do with how Windows measures sizes. Here is a snip of a folder I keep, well see the title, things in. The Size and Size on Disk are different as are all files.
Are both drivse formatted with the default sector size of 4096K?

I think you mean allocation unit size/cluster size. You can't change a HDD's sector size. Yes, all drives are formatted the default 4K cluster size.

Look at it this way, I have two identical 1TB HDDs 1,000,202,039,296 bytes (931GiB).

I copy 1 file onto each HDD, the size of that file is 31GiB. When I check Drive Properties it should show the exact same amount of Used space, which is 31GiB.

The problem I'm having is that it doesn't show the exact same amount of Used space. One drive is showing 31GiB used, and the second drive is showing 32GiB used.

 

[Britton30]

Yup, sorry I did mean allocation unit size. Since one is WD, the other Hitachi, it may be some glitch in the drives' controllers and how data is handled, the newer being more efficient...or something.
It's a curiosity though, but if all is working well it shouldn't cause any issues.

Do both drives have the same SATA interface, SATA SATAII, SATAIII, etc?

 

[0pTicaL]

The WD Black is SATA2 and the HGST is SATA3.

I guess this is one of those weird anomalies I'll never know the cause to.

 

[Britton30]

IT shouldn't make a difference but I wonder how they'd read if connected to your motherboard? On the correct SATA II or SATA III ports.
This is just a WAG (wild-ass guess) that the two SATA interfaces read bits and bytes differently.

 

[LMiller7]

The only thing I can think of is the Hitachi uses Advanced Format and the WD does not, does AF affect available storage space?

I would suspect yes.

In a standard format disk the sector size is 512 bytes while with advanced format it becomes 4096 bytes, the same size as a standard cluster. Thus, advanced format shouldn't effect file storage efficiency. But the NTFS metadata, data used by the file system itself, may well be effected. Much of the internal operation of NTFS has never been been publicly documented.

 

[TVeblen]

Another suggestion would be to use Partition Wizard (you can use the bootable CD if you don't need/want to install it on your system). See if it reports unallocated space that Windows does not report.
Free download Magic Partition Manager Software, partition magic alternative, free partition magic, partition magic Windows 7 and server partition software - Partition Wizard Online

 

[0pTicaL]

IT shouldn't make a difference but I wonder how they'd read if connected to your motherboard? On the correct SATA II or SATA III ports.
This is just a WAG (wild-ass guess) that the two SATA interfaces read bits and bytes differently.

PATA, SATA, SAS are all controllers and SATA I-III are different generations of the same controller. I haven't seen evidence indicating that one controller would write more data to disk than the other when writing a file of the exact same size.

I would suspect yes.

In a standard format disk the sector size is 512 bytes while with advanced format it becomes 4096 bytes, the same size as a standard cluster. Thus, advanced format shouldn't effect file storage efficiency. But the NTFS metadata, data used by the file system itself, may well be effected. Much of the internal operation of NTFS has never been been publicly documented.

That was my initial impression because I remember reading a long article when Advanced Format was introduced. There was something about how NTFS interprets Legacy vs AF architecture but I forgot if that had anything to do with how data is written to the drive.

Another suggestion would be to use Partition Wizard (you can use the bootable CD if you don't need/want to install it on your system). See if it reports unallocated space that Windows does not report.
Free download Magic Partition Manager Software, partition magic alternative, free partition magic, partition magic Windows 7 and server partition software - Partition Wizard Online

I don't think that would be the case because initially the difference was 1.05GiB. After I deleted the Hitachi volume and quick formatted the drive and copied the data over again from the WD the difference is now 40.2MB.

Like what LMiller7 said, the difference could just be metadata left over on the WD drive since that drive is a 2009 drive while the Hitachi is a 2012 drive and Advanced Format did not hit the market until 2010.

 

[Britton30]

No Offense, but thinking something is so doesn't make it so. Testing as TVeblen or swapping controllers in a PC as I suggested may or may not shed a light.

 

[0pTicaL]

No Offense, but thinking something is so doesn't make it so. Testing as TVeblen or swapping controllers in a PC as I suggested may or may not shed a light.

None taken and you may be right, but if a specific type or generation controller caused more data to be written wouldn't there be articles all over the web indicating such issues? I think people would be up in arms if a certain controller caused more data to be written than normally should.

eSATA is currently connected to a Marvell controller, I'm going to connect it to the Intel 6Gb/s (Z68) and see if that makes any difference.

Thinking back, I forgot to mention the EX-503 enclosure was initially connected via USB 3.0 (Renesas controller) and I switched to eSATA because transferring files over USB 3.0 would cause folders I was opening on the external to lock up, but the transfer would still continue in the background and transfer speed was awfully slow (~26MB/s).

This new enclosure is giving me a lot headaches.

 

[Britton30]

As I said, it's a WAG and may have nothing to do with the size discrepancies. I'm a tinkerer and fiddler-arounder and try all sorts of (sometimes crazy) things.

I know Marvel SATA is slower than Intel's Z68 one though. eSATA should be 20-30 or more times faster than USB3.0 even though it's suppose to have a through put of 5Gbs. eSATA is like an extension cord to a board SATA port.

 

[0pTicaL]

When I first got my EX-503 I popped the back open to replace the stock fan with a quieter 140mm fan. I noticed the board was just a plain jane eSATA to USB 3.0 board but I decided to connect it via USB 3.0 anyway because I wanted to keep all the cables in the back and I knew the 2x eSATA ports in the back were crappy jmicron ports (I just didn't know how crappy).

When connected through USB 3.0 I would get lock ups and BSODs that's when I had to switch to eSATA. I started with the eSATA the ports in the back (P8Z68 Deluxe) and that's when I found out what garbage of a controller jmicron is on this mobo. It's worse than Marvell's controller and I was pulling around ~30MB/s on sustained transfers. Then I switched to the front eSATA port (Marvell) sustained speeds improved a little but not by much ~55MB/s.

Finally I switched the front eSATA to Intel's 6Gbps port (I know 3/6Gbps doesn't make a difference for HDDs and I only have 1 disk that requires 6Gbps anyway, my 256GB Vector) speeds were around 80MB/s sustained, but still slow ass heck when moving around 2TB of data.

I might just end up getting a dedicated eSATA card because I really want to keep all the cables in the back but I think a good card runs around $300.