[SOLVED] same WD external hdd's - different speeds

nickbeef

Distinguished
Jul 15, 2015
54
1
18,545
I have two separate Western Digital Passport external drives that I use for temp storing and backing up of media files when I'm working in the field (I'm a video producer, so I need to offload and backup camera and audio files before storing them to my home RAID). Recently I've been shooting RAW video which has very large files sizes (1Gb/s - 128GB cards fill after 30 mins of recording time), and so dumping data off the cards take a long time.

I noticed that one of my drives was taking longer to offload cards onto it compared to the other drive - even though the amount of data on the cards was comparable. So when I got home I ran CrystalDiskMark on both drives and this is what I got:

Drive A
SeqQ32t1 - READ - 34.15 MB/s
Drive B
SeqQ32t1 - READ - 119.5 MB/s


Drive A
SeqQ32t1 - WRITE - 35.04 MB/s
Drive B
SeqQ32t1 - WRITE - 104.7 MB/s

The difference seems significant. Otherwise, these should be the exact same drives.
 
Solution
Those differences seem to indicate drive A running at USB 2.0 speeds, while drive B is at USB 3.0 speeds.
That was my thinking, too. Since USB 3.0 uses different pins in the cable than 2.0, it could be that one or more of the 3.0 data pins are open circuit. It would be interesting to test the faster drive on a USB 2.0 port to see if the lower speeds are replicated.

popatim

Titan
Moderator
Run a crystaldiskinfo and check SMART output too. Check the reallocated and uncorrectable sectors results, anything other then a 0 in the raw section is an issue.

Also, how full the drive is makes a difference too. Spinning disks slow dows as the heads move in towards the center due to the platters moving slower beneath the heads.

Next up, imo, would be write caching. Enabled is faster.
run device manager and right click the disk and select properties. You can find caching on the policy tab.

Check the format and cluster size of each drive.
There is a noticeable speed difference between 512 sectors and 4k+ sector size.
open and admin cmd prompt, type in: chkdsk drive:
ex: chkdsk s:

This will run a quick chkdsk and will display the allocation units.
 

nickbeef

Distinguished
Jul 15, 2015
54
1
18,545
Run a crystaldiskinfo and check SMART output too. Check the reallocated and uncorrectable sectors results, anything other then a 0 in the raw section is an issue.
Ok, I did this. CrystalDiskInfo has "Helth Status" as "Good" on both external drives (though one of my internal drives is "Caution".) Both external drives have "00" in both reallocated and uncorrectable sectors (my internal drive has "006" in Uncorrectable Sectors)

Also, how full the drive is makes a difference too. Spinning disks slow dows as the heads move in towards the center due to the platters moving slower beneath the heads.
Both are the same 4TB drives. The slow drive is a little bit fuller then the faster one, but both are quite full and going to get fuller. The only thing on the drives are a current video project that I'm shooting, but it's in RAW, so it's going to fill the drives. Slow drives has 465GB out of 3.63TB free. Faster drive as 609GB out of 3.63TB.

Next up, imo, would be write caching. Enabled is faster.
run device manager and right click the disk and select properties. You can find caching on the policy tab.
Already done.

Check the format and cluster size of each drive.
There is a noticeable speed difference between 512 sectors and 4k+ sector size.
open and admin cmd prompt, type in: chkdsk drive:
ex: chkdsk s:
Did this but don't quite know how to read the data. But both look similar, both are NTFS, and both read "976745727 total allocation units on disk"

This will run a quick chkdsk and will display the allocation units.
 
Those differences seem to indicate drive A running at USB 2.0 speeds, while drive B is at USB 3.0 speeds.
That was my thinking, too. Since USB 3.0 uses different pins in the cable than 2.0, it could be that one or more of the 3.0 data pins are open circuit. It would be interesting to test the faster drive on a USB 2.0 port to see if the lower speeds are replicated.
 
Solution