Home · Register · Software · Software · Join Upload & Sell

Moderated by: Fred Miranda
Username  

  New fredmiranda.com Mobile Site
  New Feature: SMS Notification alert
  New Feature: Buy & Sell Watchlist
  

FM Forums | Nikon Forum | Join Upload & Sell

  

XQD Reader testing
  
 
reggieb
Offline
• •
Upload & Sell: Off
p.1 #1 · p.1 #1 · XQD Reader testing


So, I've seen some tests of various XQD cards, but not of the readers (NOTE: since starting this, I found I am wrong, but maybe this will still be useful info - cameramemoryspeed.com does provide reader info).

When I ordered my first XQD cards, I didn't realize it would come with a reader, so I also ordered a reader. I now own more XQD readers than cards, but I own different ones, so I decided I would compare them and see what would happen. Note that I didn't do a TON of testing, 1 file copied to, then back off of, 2 different cards in each of the 3 reader types that I own. I own a few Sony G-readers, but I only tested one of them.

Methodology

  1. Before each test, I reformatted the XQD cards in exFAT. The reason that I chose this file system, though it is different than it will be when formatted for your camera, but I was using a single file that was too large for the FAT32 file system.

  2. The file that was copied was a 4.65 GB archive file.

  3. I ran through each of the 3 card readers with a Sony G-card, then a Lexar 2933x card.

  4. I used the Google Search stopwatch to time, and monitored each transfer using the Windows explorer file manager status window. Note that this means that the times aren't exact, I shot over to the timer as quickly as I could, but there's probably about a +/- 2-3 second error for each recorded time


The 2 cards used were:
1. 32 GB Lexar 2933x card (rated to 440 MB/s)
2. 32 GB Sony G card (the older 400 MB/s version)

The 3 readers used were:
1. Sony MRWE80 reader, this is the one that I purchased separately. I figured it was slower than the XQD 2.0 readers that I use, but it reads all of my cards without issue, including my Sony S cards. So that's what I've always used. This is an old reader that is no longer available.
2. Sony "XQD USB Adapter for G series" (QDA-SB1A) that came with my G cards that I got from Lens Authority. The model number for this one is QDA-SB1A
3. Lexar LRWXQDU-7000 Rev A this reader came with the 2933x Lexar card.

My first test (I plan to do more tests - I'll use my personal desktop and a group of D850 files - as I am renting it for an event this weekend, and update this over the weekend) involved writing to and reading from the internal SSD on my HP laptop (provided for my day job by my employer).

Here are my raw results, followed by some notes on the subject.

Sony MRWE80
Write speeds:
Sony - 1:24 (55.4 MB/s)
Lexar - 1:27 (53.4 MB/s)

Read speeds:
Sony - 0:59 (78.8 MB/s)
Lexar - 0:38 (122.4 MB/s)


Sony QDA-SB1A:
Write speeds:
Sony - 1:32 (50.5 MB/s)
Lexar - 1:00 (77.5 MB/s) *Note the time on this that I got was actually 59.97s - the only write time under 1 minute)

Read speeds:
Sony - 0:47 (98.9 MB/s)
Lexar - 0:38 (122.4 MB/s)


Lexar LRWXQDU-7000 Rev A
Write Speeds:
Sony - 1:16 (61.2 MB/s)
Lexar - 1:11 (65.5 MB/s)

Read speeds:
Sony - 0:48 (96.9 MB/s)
Lexar - 0:35 (132.9 MB/s)


A couple of interesting notes:


  1. The Lexar obtained its average speeds by having much higher peak speeds, but a pretty volatile speed with big spikes and troughs. The Sony readers were much more consistent during transfers, though the peaks were a lot lower

  2. I find it really interesting that the Sony card did best with the Lexar reader, and the Lexar card did best with the Sony G card reader.

  3. All of these speeds are a lot slower than published speeds. I'd like to test out the Sony XQD/SD reader that looks like it's quite a bit faster than the other card readers tested at cameramemoryspeed.com, but I also need to try on a different computer, I think that the controller on this laptop could be a bottleneck. Though it does have an SSD, the computer itself is really slow on the whole.


Hope that someone finds this useful. Also, if you need an XQD reader, I have readers that aren't even out of their packaging because I have so many of the things.



Sep 29, 2017 at 03:32 PM
CanadaMark
Online
• • • •
Upload & Sell: Off
p.1 #2 · p.1 #2 · XQD Reader testing


I see faster read speeds from my 4 year old CF cards (rated for 160MB/s) - I wonder if something else is going on. Motherboard and destination hard drive can also make a big difference - what components exactly were you transferring to? It looks like you may be limited by one of the two which often cap out around your reported maximums, even with some lower-end SSDs which in some laptops can be barely any faster than a HDD. Same thing with the stability of the transfer (you mention it jumps around a lot), that is likely an issue on the computer side.

Your results look like what they would have on my old PC, which were highly variable transfer maximums and generally slower overall speeds. On my new PC, everything is pegged at the maximum with no slowdown or significant variation for the entire duration of the transfer - a lot has to do with the motherboard and destination drive.



Sep 29, 2017 at 04:05 PM
gdsf2
Offline
• • •
Upload & Sell: Off
p.1 #3 · p.1 #3 · XQD Reader testing


Did you use a USB 3 Port on the PC. Any other read or write activity going on?


Sep 29, 2017 at 04:13 PM
CanadaMark
Online
• • • •
Upload & Sell: Off
p.1 #4 · p.1 #4 · XQD Reader testing


gdsf2 wrote:
Did you use a USB 3 Port on the PC. Any other read or write activity going on?


He had to be using USB 3.0 if he got more than about 60 MB/s.

Also, to your point, if he was copying to the OS drive (where there is always read/write activity) the test was not ideal. A secondary, fresh NVMe/PCI SSD with write speeds well above the cards maximums would be the only way to properly test. You would also want to be using a modern, high end motherboard and plugged physically into one of the mobo USB 3.0 ports (not one of the case extensions) or a PCI-E USB 3.0 adapter.



Sep 29, 2017 at 04:25 PM
elkhornsun
Offline
• • • •
Upload & Sell: Off
p.1 #5 · p.1 #5 · XQD Reader testing


I do not recommend using the old Sony Sony MRWE80 reader. I found it has a bad driver and so the Windows file manager would often not recognize the reader or a XQD card when inserted. Then I was having boot problems with the computer and it took awhile to realize that it was the Sony card reader that was causing the Intel BIOS to think that it was to boot from an external network drive. No such problems with my Lexar card readers or any other in the past but then it is one aspect of the Windows design that goes back to NT 3.51 where the kernel is directly exposed to COM ports and so inserting a USB device can crash the operating system.

I am having no problems at all with the Sony QDA-SB1A card adapter and that is what I would recommend people buy to connect to their XQD cards.



Sep 29, 2017 at 04:32 PM
CanadaMark
Online
• • • •
Upload & Sell: Off
p.1 #6 · p.1 #6 · XQD Reader testing


elkhornsun wrote:
Then I was having boot problems with the computer and it took awhile to realize that it was the Sony card reader that was causing the Intel BIOS to think that it was to boot from an external network drive.


Check your boot order, you can prevent that.



Sep 29, 2017 at 04:45 PM
 

Search in Used Dept. 

        


reggieb
Offline
• •
Upload & Sell: Off
p.1 #7 · p.1 #7 · XQD Reader testing


CanadaMark wrote:
He had to be using USB 3.0 if he got more than about 60 MB/s.

Also, to your point, if he was copying to the OS drive (where there is always read/write activity) the test was not ideal. A secondary, fresh NVMe/PCI SSD with write speeds well above the cards maximums would be the only way to properly test. You would also want to be using a modern, high end motherboard and plugged physically into one of the mobo USB 3.0 ports (not one of the case extensions) or a PCI-E USB 3.0 adapter.


As I mentioned, I am pretty sure that my low speeds were due to the computer I was transferring to. And I don't use a secondary fresh NVMe/PCI SSD, so the test isn't supposed to be a "How fast can it work theoretically" so much as "How fast are they in practical use." But yeah, the PC is why the overall speeds were slow I think.



Sep 29, 2017 at 05:51 PM
sjms
Offline
• • • • • • •
Upload & Sell: On
p.1 #8 · p.1 #8 · XQD Reader testing


I was about to mention that the computer and what's going on will have a great effect on that xfr

Before each test, I reformatted the XQD cards in exFAT. The reason that I chose this file system, though it is different than it will be when formatted for your camera, but I was using a single file that was too large for the FAT32 file system.

this skews real world results due to the fact that the cameras themselves use FAT32 and with all its pluses and minuses. practicality dictates that you use the system as it would be in actual use. in essence you did exactly what you didn't want to do. its those "limitations" that make it flow the way it does.

using an all real world environment with repeatable results I averaged going from card to desktop 265MB/s 135 14 bit compressed raw images shot today using the Sony MRW-E90 dual reader in a MB based USB3 port


Edited on Oct 01, 2017 at 01:36 PM · View previous versions



Sep 29, 2017 at 10:10 PM
EB-1
Offline
• • • • • • •
Upload & Sell: Off
p.1 #9 · p.1 #9 · XQD Reader testing


gdsf2 wrote:
Did you use a USB 3 Port on the PC. Any other read or write activity going on?

CanadaMark wrote:
He had to be using USB 3.0 if he got more than about 60 MB/s.

Also, to your point, if he was copying to the OS drive (where there is always read/write activity) the test was not ideal. A secondary, fresh NVMe/PCI SSD with write speeds well above the cards maximums would be the only way to properly test. You would also want to be using a modern, high end motherboard and plugged physically into one of the mobo USB 3.0 ports (not one of the case extensions) or a PCI-E USB 3.0 adapter.


We test storage devices by reading/writing the files with a utility rather than adding a variable of copying to a second storage subsystem.

EBH



Sep 30, 2017 at 04:49 AM
Charles Loy
Offline
• • • •
Upload & Sell: On
p.1 #10 · p.1 #10 · XQD Reader testing


@ reggieb - I found this interesting and useful. I have the two Sony cards you tested. Thanks


Oct 01, 2017 at 11:16 AM
sjms
Offline
• • • • • • •
Upload & Sell: On
p.1 #11 · p.1 #11 · XQD Reader testing


CanadaMark wrote:
He had to be using USB 3.0 if he got more than about 60 MB/s.

Also, to your point, if he was copying to the OS drive (where there is always read/write activity) the test was not ideal. A secondary, fresh NVMe/PCI SSD with write speeds well above the cards maximums would be the only way to properly test. You would also want to be using a modern, high end motherboard and plugged physically into one of the mobo USB 3.0 ports (not one of the case extensions) or a PCI-E USB 3.0 adapter.


I copy direct to the desktop first via drag and drop then to my Drobo 5D same method after folder is renamed all done through USB3. any informational additions/conversions are done later.




Oct 01, 2017 at 12:28 PM







FM Forums | Nikon Forum | Join Upload & Sell

    
 

You are not logged in. Login or Register

Username     Reset password