PhotoShelter Card Reader Database & Why You Should Upgrade

PhotoShelter Card Reader Database & Why You Should Upgrade

Of the myriad ways to waste time (e.g. incessantly checking Facebook or Instagram, marathon viewings of Game of Thrones), waiting to ingest your digital photos probably doesn’t register on your radar. Nevertheless, if you’re using an old or poor performing card reader, you’re unnecessarily losing hours of your life.

As you can probably surmise, not all card readers are the same. Some have better components and faster chips, which results in significant time savings. Quantifying this time was our key goal in testing a number of card readers on the market.

In trying to update Rob Galbraith’s seminal CF/SD/XQD database for current card readers, one thing became eminently clear: technology has progressed rapidly, and what was once considered fast, is not even remotely true anymore.

With this in mind, we didn’t bother to test USB 2.0 devices or anything slower than a Class 10 media because you’re wasting time and not using your expensive gear to its full potential if you haven’t upgraded your cards and readers in the past two years or so.

The Test Systems:

Late 2013 Mac Pro

  • 3.5 Ghz 6-Core Intel Xeon E5
  • 64GB 1867 Mhz DDR3 ECC
  • OS X 10.9.4
  • 1TB Flash Storage

Mid 2014 MacBook Pro

  • 3.0 Ghz Dual-core Intel Core i7
  • 16GB 1600 Mhz DDR3L SDRAM
  • OS X 10.9.4
  • 512GB Flash Storage

Card Readers

How do I know if I have a USB 3.0 card reader?

All the USB 3.0 card readers we tested had the USB 3.0 Micro-B plug, which is characterized by the dual-sectioned form factor.

Connector_USB_3_IMGP6033_wp

Photo: Rainer Knäpper, Free Art License (http://artlibre.org/licence/lal/en/)

Methodology:

The following results were generated with QuickBench 4, a cross-platform benchmarking utility that Rob Galbraith used for his original database. The cards were formatted using a Nikon D800 (or Nikon D4 for XQD cards). Prior to testing, we ran an experiment to see if a low level erase affected performance, and the difference was negligible (this bodes well for real-life performance). QuickBench was set to Large Test (2-10MB), Allow Cache Effects was turned off and 3 test cycles were run, and the average of those tests is displayed below in MB/sec.

The Results (MB/s)

“Error” indicates that the software produces no read/write cycles after 1 min. “n/a” means the device doesn’t support the card type.

Larger numbers are better.

 

 

Real life ingest speeds:

QuickBench is a good utility, but lab-style read/write cycles isn’t reality. So we tested real-life ingestion. 100 RAW photos from a Nikon D4 were transferred using the fastest/slowest card and fastest/slowest card reader combinations to illustrate how dramatic the difference in ingest duration can be. Here are the transfer times using Photo Mechanic v5 on the MacPro.

Transfer time (in seconds) with the fastest CF reader (Lexar):

  • SanDisk Extreme 32GB CF: 14.35s
  • Transcend 16GB CF: 14.66s

Transfer time (in seconds) with the slowest CF reader (Sonnet):

  • SanDisk Extreme 32GB CF: 77.16s
  • Transcend 16GB CF: 176.38s

Transfer time (in seconds) with the fastest XQD reader (Sony):

  • Sony XQD S Series: 10.66s

As you can see, the difference between the fastest card/reader and slowest card/reader combination to transfer 100 RAWs is 162s (2min 42s for the arithemetically challenged).  Now extrapolate that to a wedding or sporting event where you might be ingesting 2,000 images, and we’re talking close to an hour difference in ingest time alone! And remember, these are USB 3.0 readers with the latest cards writing to a solid state disk. If you have USB 2.0 and slower generation cards writing to a platter-based hard drive, you’ll have enough time to watch an entire season of GOT.

Key Takeaways:

  • The difference between the fastest card+fastest reader and the slowest card+slowest reader is significant.
  • If your computer only has USB 2.0 or FireWire 800 ports, you are barely scratching the surface of the potential read speeds.
  • UPGRADE YOUR DAMN CARDS AND READERS NOW

Insofar as ingest time is concerned, the type of card matters less than the type of reader. Given the relatively low cost of the fastest readers (Lexar, Kingston, SanDisk), it’s an absolute no-brainer to update your card reader immediately.

Next Post:
Previous Post:
This article was written by

Allen Murabayashi is the co-founder of PhotoShelter.

There are 20 comments for this article
  1. Pingback: PhotoShelter Card Reader Database & Why You Should Upgrade | PhotoShelter Blog – The Click
  2. Kent at 10:12 pm

    Cool. thanks for sharing Allen. That being said I don’t see the Lexar Professional Workflow CFR1 card reader on here. Definitely one of the best I’ve used. It doesn’t have the mini-usb3 plug, uses the full size one. One problem I kept running into was that the miniusb3 plug kept wanting to come out.

  3. Florian at 1:14 am

    I agree with Kent, the Lexar workflow would be an interesting test. Both in USB 3 and thunderbolt, and single vs all 4 bays.
    Thanks for doing the update.

  4. Steve at 1:29 am

    Thanks a bunch Allen. Some good info. Isn’t the thunderbolt sonnet adapter the fastest since its thunderbolt compared to usb 3.0?

    • Allen Murabayashi Author at 11:08 am

      That is what I was expecting, but it seems like there is some overhead with the cardexpress format that makes it slower than anticipated.

  5. Martin at 10:59 am

    Looks like these are all external card readers. As a general rule, is there any performance difference between external and internal (built-in) readers? Or is that just a difference in form factor?

  6. Brian Bohannon at 1:14 pm

    Thanks for the timely article, Allen. I recently upgraded to Nikon D810 and D750 bodies from the D700, and quickly noticed the need for newer/better/faster cards, reader and computer! My local camera store didn’t know what to tell me about readers, so thanks for doing my homework for me. Went with the Lexar Pro card reader and a mix of Sandisk and Lexar cards, which I’ve had good results with in the past.

    Doing my part for B&H and the economy …

  7. Joe at 10:52 am

    – You missed testing PNY SD cards. Sold at a great price. PNY is HPs brand. While not as blindingly fast as top line pro by SanDisk, the price to speed ratio is impressive. They are also crush, dust, and water proof.
    – On the oposite of the price spectrum is Hoodman SD memory, which has taken top spot on other comparisons (but carries a price to match its top performance.)
    – Wondering: how does reading from camera compare? Maybe also testing Nikon 610/810, Sony 99, Leica, and Cannon FF equivalents too would be good. Supposedly, Nikon bodies work faster with a specific brand. SanDisk, I think?

  8. Joe at 12:24 pm

    I could not read the document on Safari. Why not post a link to a version that is text only or, build it in HTML so it works across browsers?

  9. Pingback: IO Computer – Video Projectors » 2in1 Apple iPad SD Card Reader and USB Camera Connection Kit
  10. Global at 5:13 pm

    THE INTERNET NEEDED THIS ARTICLE! Also a D810 user. Its funny you mentioned Galbraith’s reviews — I used them when making my first USB purchases YEARS ago!! Your update is exactly what I was hoping for from him. But found yours instead. Thanks! 🙂

    Would like to see your conclusions about which reader to buy based on real-world outcomes.

    Also, the difference on a PC, if it matters.

  11. Cory at 3:08 pm

    I have found the new UHS-II SD cards to be faster than any CF cards. While there are not many cameras yet available to take advantage of their speed while shooting, they are backwards compatible and the speed advantage is useful for the card to computer transfers.

    • Allen Murabayashi Author at 3:13 pm

      We didn’t test the UHS-II because only the Fuji XT1 supported that technology at the time of our testing. But yes, it’s blazingly fast at a price.

    • Allen Murabayashi Author at 3:15 pm

      Stewart, see above in the comments. The Thunderbolt Express reader didn’t test well — seems like there’s some processing overhead when converting from expresscard/34 to thunderbolt that makes it slower than the fastest USB3 readers

  12. Brian Bohannon at 11:16 am

    So I unpacked my new Lexar Card reader and before even reading a card with it, broke it.

    You squeeze two tabs on the bottom front sides of the device, and the reader pops up to expose the two card slots. I pushed one before I found the other, and now it’s stuck. Reader will still pop up, but one tab is stuck, and reader will now pop up on it’s own in my bag …

    Janky piece of hardware.

  13. Pingback: Memory Cards – What Speed Do I Need? | Kevin Davis Photography
  14. Pingback: Tarjetas y lectores, especificaciones ténicas | Guadalupe García Pasquel

Leave a Reply

Your email address will not be published. Required fields are marked *