Kingston Technology DataTraveler USB 3.1 Two Drives But Diferent Performance

[Mark55]

Hi

Back in June I purchased a Kingston Technology Data Traveller USB 3.1 64Gb Version 1 from Amazon UK

Its Crystal Mark Result is:

Sequential Read (Q= 32,T= 1) : 50.810 MB/s
Sequential Write (Q= 32,T= 1) : 12.611 MB/s
Random Read 4KiB (Q= 32,T= 1) : 5.005 MB/s [ 1221.9 IOPS]
Random Write 4KiB (Q= 32,T= 1) : 0.127 MB/s [ 31.0 IOPS]
Sequential Read (T= 1) : 49.706 MB/s
Sequential Write (T= 1) : 12.583 MB/s
Random Read 4KiB (Q= 1,T= 1) : 4.472 MB/s [ 1091.8 IOPS]
Random Write 4KiB (Q= 1,T= 1) : 0.131 MB/s [ 32.0 IOPS]

Test : 1024 MiB [K: 0.0% (0.0/57.6 GiB)] (x1) [Interval=5 sec]
Date : 2016/01/21 0:52:19
OS : Windows 10 [10.0 Build 10586] (x64)

I have now purchased another such drive from Amazon, Version PMAP

Its Crystal Mark result is:

Sequential Read (Q= 32,T= 1) : 112.035 MB/s
Sequential Write (Q= 32,T= 1) : 13.448 MB/s
Random Read 4KiB (Q= 32,T= 1) : 11.151 MB/s [ 2722.4 IOPS]
Random Write 4KiB (Q= 32,T= 1) : 0.045 MB/s [ 11.0 IOPS]
Sequential Read (T= 1) : 116.193 MB/s
Sequential Write (T= 1) : 8.389 MB/s
Random Read 4KiB (Q= 1,T= 1) : 10.385 MB/s [ 2535.4 IOPS]
Random Write 4KiB (Q= 1,T= 1) : 0.104 MB/s [ 25.4 IOPS]

Test : 1024 MiB [J: 0.0% (0.0/57.7 GiB)] (x1) [Interval=5 sec]
Date : 2016/01/21 0:56:33
OS : Windows 10 [10.0 Build 10586] (x64)


Average write speed of the Drive purchased in June 2015 is 45MB/s (video files)

Newer drives purchased in January 2016, average write speed is 20MB/s (video files)

It seems newer does not mean faster, or even the same speed.

Why would Kingston Technology change something in the drive only to make it slower?

Thanks in advance Mark55

 

[Ralston18]

Not sure that I understand your question.

The first (June) Kingston drive has mostly slower Crystal Mark results than the latter (Amazon) drive.

You mention Version PMAP in the later but not the former. Are the drives truly identical?

The June drive is faster with the video file transfers as I understand your posting.

How did you determine the average write speed (video files) for each drive?

Are you sure that the encryption and/or indexing (if any) was identical?

Are both drives USB 2.0, USB 3.0, or one of them just USB 3.0 compatible. Meaning it will work in a USB3.0 port but only at USB 2.0 speeds. (Marketing gimmick.)

Were both drives formatted exactly the same? Did you use the same USB port, USB cable, same computer? No changes at all during the testing, no other software running or no longer running.

How many trials did you run?

I doubt that Kingston would intentionally change something to make the USB drive slower. Something else may be going on.

All in all the slower USB drive could have a problem. Quality Control is becoming more of a problem....

On the other hand I would be concerned about the testing being truly consistent.

Not disagreeing with your testing and results per se but I need some help with understanding the bigger picture. Thanks.

 

[Mark55]

Hi
Both drives are Data Traveller G4 64GB.

The older June drive is USB 3.0 and Crystal Mark reports it as Version 1.
The newer January drive is USB 3.1 and Crystal Mark reports it as Version PMAP.

Other than that the drives appear visually identical.

Testing method:

Computer running Windows 10, each drive was plugged in to the same USB 3 socket on the rear of the motherboard.

Each drive was formatted in ExFat just prior to each test.

The test involved writing several video MOV files, each around 600Mb in size. the files were stored on a Samsung SSD. I ran each test 10 times and changed the files names between each test. I only tested the write speed of each drive.

On each test the newer January drive ran 2.04 times slower than the older June 2015 drive.

However the thing that got my attention and started me testing the drives was.

The older drive write graph was almost flat and steady with the write speed varying between 44MB/s and 45MB/s

The newer drive had a very different write graph that looked like a wave form varying between 16MB/S and 22MB/s

I will see if I can post a screen shot of the above.

Just to say the testing was carried out on 2 drives from June 2015 and 2 drives from January 2016.

I also note Kingston Technology do NOT publish write or read data for these drives other than the word "Standard"

Perhaps I just got lucky with the June 2015 dives!

Thanks

 

[Ralston18]

Excellent - thank you. Nice methodology.

One of the reasons I like this forum so much is that there are always new and interesting things to learn.

I am now interested the "PMAP Version" vs "Standard".

I googled "PMAP" and found a variety of links. Here is one of them:

http://linux.die.net/man/1/pmap

There were a variety of links relating to the "PMAP" acronym but many were obviously not relevant.

But I am intrigued by the Linux aspect. Maybe the PMAP Version drive respresents some failed product intended for the Linux market and has some build in design "features" intended for future marketing.... Did not work, as you may have discovered, but they were sold anyway. (FYI, I do tend to be quite cynical sometimes.)

Anyway, normally I would pursue a bit more research before commenting or replying. However, I happen to be very near the bullseye for the incoming East Coast blizzard. Lots to do and could lose power later in the weekend. May or may not be able to get online. TBD.

Felt it best to at least share the Linux information.