REVIEW: X-Pro2 SD Memory Card Buffer test
How many images can an X-Series camera shoot before it runs out of steam? This is a question I hear a lot.
The Fujifilm X-Pro2 is the first X-Series camera to incorporate two memory card slots. The first utilises the new, fast SDHC II cards while the second accepts the slower SDHC I memory. You can fit a slower SDHC I card in slot 1 and conversely you can put a SDHC II card in the second slot but it will only work at the slower write speeds.
I have been using the faster SDHC II cards in my X-T1s since April 2014 as it speeds up the processing of the images and frees up the camera quicker after a continuous burst at 8fps.
I have long known the benefits of using the faster cards for camera operation and last weekend on the Snowman Rally I was shooting with the X-Pro2 on 8fps and the cards I am using certainly enabled me to continuing shooting uninterrupted.
Anyway a lot has been written about the faster AF, the faster processing speed etc etc of the X-Pro2 but I wanted to conduct a quick, and very unscientific test, to find out how many shots I can take with the camera shooting at 8fps before the memory card would buffer.
For this test I am using six different cards, two are SDHC I and four are the newer SDHC II type cards. If you flip the card over you can see the extra set of pins on the type II cards compared to the older type I.
I am currently using a range of SD cards across my X Series cameras -
- Sandisk Ultra 30MB/s SDHC I 16GB
- PNY Elite Performance 100 MB/s SDHC I 32GB
- LEXAR Professional 1000x 150MB/s SDHC II 32GB
- Toshiba Exceria Pro 240 MB/s SDHC II 16GB
- Sandisk Extreme Pro 280 MB/s SDHC II 32GB
- LEXAR Professional 2000x 300 MB/s SDHC II 32GB
To conduct the test I fitted the 18mm f2 lens to the X-Pro2 and set the camera to manual exposure (1/250s @ f2 ISO1600), set the focus to manual and the drive to continuous High (8fps) . The first round of testing was conducted with the camera set to record RAW + JPEG N as this is the mode I tend to shoot in when out working.
I then fired the shutter until it stopped. I immediately removed my finger from the shutter release and timed how long it took for the light on the back of the camera to stop blinking, which indicated that the buffer had cleared. I also conducted this test in slot 1 (the faster slot) and then switched the card to slot 2 and retested. I formatted the card between each test so it was clear of any images.
The first card tested was the slowest, a Sandisk Ultra 30 MB/s SDHC I 16GB card. Unsurprisingly this was the slowest of the group but the performance was way off the pace. The camera would shoot 24 shots before the cameras buffer was full but the camera was then tied up for nearly 150 seconds before the buffer was cleared. Switching the card to Slot 2 gave the same performance.
I then moved on to the PNY Elite Performance 100MB/s SDHC I 32GB card. Things were a lot better with this card in slot 1. The camera took 25 shots before the buffer was full but more significantly only took 24.3 seconds to clear the buffer. The camera was able to shoot again after a few seconds as buffer space became available. The performance was exactly the same in Slot 2 of the camera.
Moving to the slowest of the SDHC II cards, the LEXAR Professional 1000x 150MB/s 32GB, I started to see a difference between the two slots as Slot 1 was able to use the higher write speeds to clear the buffer faster. The buffer was filled after 25 shots but only took 18.5 seconds to fully clear the buffer. Once again the camera was in operation after just a few seconds as the buffer was cleared as the images were written to the memory card. In Slot2 the LEXAR gave the same 25 shots and took 2.5 seconds longer to clear the buffer.
Next up was the Toshiba Exceria Pro and this was the first card where I saw an increase in the number of shots before the camera buffer was filled. With the faster write speeds the camera was able to transfer images out of the buffer before it was filled so I reached 27 shots, two more than the LEXAR, before the buffer was full. It only took 12.9 seconds before the buffer was cleared, 5.6 seconds quicker than the 150 MB/s LEXAR. In slot 2 the camera buffered at 25 shots and took 17.6 seconds to clear.
I expected great things from the Sandisk Extreme Pro 280 MB/s card but in reality it was actually slower than the Toshiba. It took the same 27 shots before the buffer was full but took 13.8 seconds to clear. In slot 2 this card took just 24 shots and a mammoth 31.6 seconds to clear the buffer, slower than all the other cards with the exception of the 30 MB/s Sandisk. I ran the test three times and the results were the same.
The final card on this test is the new LEXAR Professional 2000x 300 MB/s and this card did not disappoint. It took 30 shots before the camera buffered in slot 1 and took just 11.5 seconds to clear the buffer. In slot 2 is took 27 shots and was clear again in just 16.8s.
JPEG only shooting
As a final test I put each card in slot 1 but set the camera to only shoot JPEG images to see how many shots the camera would take. The results were quite impressive with the slowest card taking 65 shots, the 100 MB/s PNY taking 120 shots, the LEXAR 1000x taking 182 images and all three of the fastest cards not buffering even when the camera passed 300 images, so I have classed these as unlimited.
UPDATE - I have just carried out the same tests on the X-T1 and you can review the results HERE
Conclusion
In reality when shooting sport or wildlife you are only going to shoot short bursts and rarely anything more than 8-10 shots. What this unscientific test shows that the faster cards do speed up camera operation and are well worth investing in to get the best out of the new breed of X Series cameras. If you are running an X-T1 or X-Pro2 I recommend buying the faster LEXAR or Toshiba SDHC II cards.
The 32GB LEXAR 2000x was £47 on Amazon and represents great value for money. I have four LEXAR memory cards and I have never had one fail on me. The same goes for Sandisk and Toshiba and PNY but I tend to buy new cards every 12-18 months to cut down on the possibility of failure and losing precious images.
A final advantage of using the faster cards is the download speed from the card to your computer, as long as you are using the latest card reader to do the transfer of course.
-------------------------------------------------------------------------------------------------------------
PLEASE SUPPORT THIS BLOG BY VISITING THE ADVERTISERS ON THIS PAGE
By clicking on the adverts you are helping support this blog - thank you.
-----------------------------------------------------------------------------------------------------------------
ALL IMAGES ARE THE PROPERTY OF MACLEAN PHOTOGRAPHIC AND CANNOT BE USED FOR ANY PURPOSE WITHOUT PRIOR PERMISSION
MacLean Photographic run Tours and Workshops in East Lothian and the Borders of Scotland. CLICK HERE for more details and availability
In June 2015 Jeff Carter was named as a Fujifilm brand ambassador and you can view his profile and gallery on the Fujifilm website HERE
If you like what you see on this blog please visit our Facebook page and click 'like'
I would've liked to see how the cards performed just in slot 1. It seems like slot 2 would be the bottleneck and why all the cards were so similar.
ReplyDeleteCarl, First of all thank you for taking the time to read and comment. However I am not sure by what you mean in by saying you'd prefer to see how the cards performed JUST in slot 1? The cards were all tested in slot 1 independent of slot 2. If you just want to know how the cards performed in slot 1 just read the slot 1 columns in the table above and ignore the slot 2 results.
DeleteThe cards tested are ones that I have bought over the years. The fastest three cards are all similar on paper but the actual results show that the Sandisk Extreme Pro is actually slower than the 'slower' Toshiba. I don't buy SDHC I cards now as all bar my X100 can utilise the new SDHC II cards and even then I see a marked increase in performance in the X100 when using the faster cards compared to the slower 30MB/s SD cards. I only keep these now as an emergency back up or to transfer files between computers in the office.
As far as slot 2 being the bottleneck, unless you shoot 30+ frames in a continuous burst regularly the X-Pro2 clears the buffer so quickly I doubt you'd see any bottleneck. These tests were at the most extreme, I never envisage getting to the stage where the camera stops due to the buffer being full even when shoot sport.
I hope this helps answer your questions?
Thanks for the tests and sharing your results and insight.
DeleteI had researched SDHC II cards in the past and from other tests and the Lexar 2000x 300MB/s card stood out really well for value, quality and performance so I purchased 1 in preparation for the X-Pro2 I ordered.
Question: I was under the impression that you would not necessarily reap that much benefit from using an SDHC II card in a slower SDHC I card slot.. not enough to possibly justify the price for the SDHC II card. But, if I'm reading this article correctly, you are saying you will still get better performance using a SDHC II card in Slot 2 even though you probably are not utilizing the SDHC II card's capabilities fully.
It that's what you're suggesting, then it's most likely I will get a 2nd Lexar SDHC II card for Slot 2 just to optimize every aspect of writing speed possible with the X-Pro2.
I highly doubt I'll be shooting more than 10 frames continuously (more likely half of that at most if I ever do). Even if I am shooting single frames in rapid succession (not burst mode, but just pressing shutter release quickly in rapid succession), I plan to write RAW+JPEG with RAW to Slot 1 and JPEG to Slot 2, and I am hoping the new X Processor Pro in the x-pro2 will be so fast I can do this and quickly check images without waiting for the buffer to clear.. and nearly all (if not all) reviews so far have suggested there should be no problem working at a very brisk pace.
Wing Yip - Thank you for reading and commenting.
DeleteYou won't be getting the full benefit of using a SDHC II card in slot 2 but as you can see from the write times in the table above there is still a performance advantage over using a SDHC I card in slot 2. Plus the fact you can swap between the cards between the slots and not worry if you have two cards the same.
If you are writing RAWS to slot 1 and JPEGS to slot 2 then I doubt you will see any fall off in performance even if you run the camera at 8 fps continuously. The X-Pro2's processor copes really well with processing the files quickly to keep you shooting. I am seriously impressed with this camera.
Thanks for the reply.
DeleteLooking forward to finally trying this out for myself soon.. provided B&H got their X-Pro2 inventory for all the earliest pre-orders fulfilled for shipment this Thursday.