13 Flash Memory Card Roundup: Does High Speed Really Matter?

Posted December 19, 2012 by Sam Chen in Reviews
Navigation:

Conclusions

memory-card-roundup-2012-custom-pc-review-6Alright, so let’s cut to the chase and answer the original question that we started out with. Do high speed memory cards really matter?

Well, yes and no. Let’s start with the yes. In our Nikon D800 continuous shooting test, a number of bottom tier Class 10 SD cards such as the Transcend Class 10 SD card and the Patriot LX Class 10 SD card failed to keep up with the upper tier memory cards. After the cache on the D800 filled up, the camera took significantly longer to offload the contents of its cache into the slower tier memory cards. Those who do a lot of sports, action or editorial (paparazzis!) photography may find that upgrading to a faster memory to be beneficial  Now, this may not apply to every camera, since the D800 is a top tier professional camera outfitted with a full-frame 36.3MP sensor. Most cameras today (even DSLRs) will come equipped with a sensor rated  at 24MP or less, which will result in significantly smaller photos, and consequently significantly shorter cache to memory card offload times as well.

Another advantage of having a fast memory card is transfer speeds. Taking a look at our sequential read/write testing, the top tier memory cards showed significantly higher transfer speeds than their lower tier cousins, which may make a huge difference for those who regularly fill up their memory cards and need to offload data constantly on to their computer. In my testing of the Lexar Professional 1000x CF card, it was capable of 143.3 MB/s sequential read speeds, which is nearly double the sequential read speed of the SanDisk Extreme 400x CF card. Of course there’s also the price to consider. The Lexar Professional 1000x costs about 1.5x the price of the SanDisk Extreme 400x CF card, so it’ll really depend on how much the user values transfer rate as the difference between the two cards in our D800 burst shoot testing was very minor.

memory-card-roundup-2012-custom-pc-review-3

With that said, let’s talk about the no. The D800, along with most newer DSLRs on the market, have very large internal memory caches (15-25 RAW files). This means that for most photographers out there, even the slower, cheaper memory cards will work perfectly fine given that you don’t need to do continuous shooting that can over-fill the internal cache. Additionally, in our real world testing, the D800 was not able to properly utilize the additional performance offered by the faster cards such as the Kingston Ultimate 600X CF, the SanDisk Extreme Pro 600X CF, and the Lexar Professional 1000X CF versus cards in the mid tier such as the SanDisk Extreme 400X CF or even the Patriot EP Class 10 SD card. Those who can afford to wait an extra minute or two, but still want the maximum performance possible out of the camera may be perfectly fine with a mid tier memory card.

Type Speed Rating Capacity Price Price/GB Purchase Link
ADATA SD Class 6 16GB $23.29 $1.46 Click Here
Patriot LX SD Class 10 16GB $14.98 $1.07 Click Here
Transcend SD Class 10 16GB $12.30 $0.77 Click Here
Kingston SD Class 10 64GB $48.74 $0.76 Click Here
Kingston Ultimate XX SD Class 10/UHS-I 32GB $82.20 $2.57 Click Here
SanDisk Extreme SD Class 10/UHS-I 32GB $31.81 $0.99 Click Here
SanDisk Extreme Pro SD Class 10/UHS-I 16GB $39.99 $2.50 Click Here
Patriot EP SD Class 10/UHS-I 32GB $27.47 $0.86 Click Here
Patriot EP Pro SD Class 10/UHS-I 32GB $52.86 $1.65 Click Here
SanDisk Extreme CF 400x/UDMA 5 32GB $86.99 $2.72 Click Here
Kingston Ultimate CF 600x/UDMA 6 32GB $70.74 $2.21 Click Here
SanDisk Extreme Pro CF 600x/UDMA 6 32GB $140.00 $4.38 Click Here
Lexar Professional CF 1000x/UDMA 7 16GB $74.95 $4.68 Click Here

Finally, let’s also take a look at the pricing of the memory cards tested today. As expected, the blazing Lexar Professional 1000x CF card came in at the highest price per gigabyte coming in at a massive $4.68/GB, while the lowest priced card was actually quite surprisingly the Kingston Class 10 64GB SD card, which came in at just $0.76/GB.

For SD cards, the best bargain (price/performance) to be found is the SanDisk Extreme SD card, which is capable of providing excellent performance at a fairly low cost of only $0.99/GB, only $0.23/GB more than the least expensive Kingston Class 10 64GB SD card, but in real world testing is capable of an additional 11 photos in a 30 second burst. In terms of the fastest SD card tested today, the award goes to the Kingston Ultimate XX SD card, which was able to best every other memory card tested today in read, write and real world performance testing. That said, special mention does go to the Patriot EP Pro SD card as it was able to come extremely close to the performance of the Kingston Ultimate XX SD card at a price point of $0.92/GB less.

For CF cards, the best bargain to be found today is the Kingston Ultimate 600x CF card, coming in at only $2.21/GB and is capable of providing very respectable performance for the price. In terms of the fastest CF card tested today, the Lexar Professional 1000x easily takes the spot, dropping a sequential read speed of 143.9 MB/s and a sequential write speed of 98.18 MB/s. It also tied with the SanDisk Extreme Pro CF card topping every other memory card in terms of real world performance, marking it for the top spot in terms of performance. Special mention goes out to the Kingston Ultimate 600x CF card for providing real world performance similar to the Lexar Professional 1000x CF card, but coming in at nearly half the price.

 

Samples provided by: Kingston, Patriot, Lexar, SanDisk


About the Author

Sam Chen

Editor-in-Chief

  • Russell Robinson

    This is a pretty good comparison, some people still think all sd cards are equal

  • Greg Zeng

    In my experiences, I’ve found flash card performance varies between individual items which seem similar: brand, model, size, speed, etc: batch & random variations.

    Speed often lessens, depending on hubs, expansion cables, and even which hardware port on my PC is used. It’d be interesting to see if bench testing within similar batch are consistent too.