Memory card hygiene

It’s not what you might think: I’m not going to be talking about washing your memory cards, or cleaning the contacts or something in a similar vein. The way you handle your memory cards – both the physical cards and the digital contents – might seem like a fairly simple and common-sense thing, but you’d be surprised at just how many risky moves I’ve seen amongst both pros and amateurs. Unsurprisingly, a lot of people were just as nonchalant about images during the film era – and back then, there were a lot more possible failure points that would result in no image, or a very poor one; this is simply because there is only ever one original copy of the image – the negative.

Fortunately, there’s no penalty for having multiple copies of a digital file; it’s the easiest thing you can do is ensure you never lose an image. Preferably, keep the copies in physically separate locations. This ensures that should one copy befall some sort of disaster, you still have another somewhere else. Until the point there is more than one copy of your file, your photographs are very much at risk. If your camera has two card slots and supports parallel writing, buy cards in matched (at least capacity and speed, if not identical duplicates) pairs and have the camera write the file to both cards every time. If not, then download the contents of your card to a computer or storage device as soon as possible once the shoot is complete; keep the original images on the card too until you make a second backup. The aim is to minimize the time when you have only one copy of the file.

Before I use a card in a camera, I will use that particular camera to format – after backing up the contents, of course – it to ensure that the file structure/ numbering etc. are fully compatible. Although almost all cameras use the same DCIM file format (hence the 8-bit filenames), they might not share the same subfolder handling structure, or worse, pick up the file numbering where one of your other cameras left off – thereby creating a mess when it comes to organization and avoiding accidentally having two or more different images competing for the same filename. Doing a format after a backup (and before the next shoot) also means you know you’ve got fresh cards and sufficient capacity remaining.

This might sound obvious, but ensure the cards are fully compatible with the camera you’re using; countless Leica M8/ M9 shooters will agree with me here. Use the wrong kind of card and lockups, file corruption and general operational bugs occur with alarming frequency. It happened to me once on a very important assignment; ever since then, I’ve made sure I religiously check the suitability of a card before shooting anything critical with it. Oddly, it’s only the Leicas which seem to have this problem.

Protect and Delete All functions
Some cameras have the ability to easily protect individual images, and/or delete multiple (or all) images. I use this to sort through my images on the fly if I happen to have some downtime during a shoot; this minimizes the amount of grading I have to do later at the computer. Of course, for this to be a workable option, you need a sufficiently high quality LCD, some sort of highlights warning and the ability to magnify to actual pixels; this way you can check focus and exposure – generally, anything that passes both gets protected for later close inspection, then I use the ‘delete all’ function to remove the remainder. It’s much faster than doing it individually, especially if there’s a way to set a single-press to zoom to pixel level and subsequently jump quickly between images to compare A-B. Note: it’s very important you do a test run with your camera to see if it handles files in the way I describe; most do (I can say with certainty that Nikon, Olympus and Leica all operate this way). If not, you may actually land up deleting every file, marked/ protected or not.

Physical handling
Though the physical design of memory cards (SD, at any rate) doesn’t appear to be that strong – they can be snapped, pins bent, or pieces of plastic simply broken off in a way that affects the connector – there are perhaps a little more robust than we might think. There are plenty of tales of people putting them in the wash and discovering they still work fine afterwards, or getting stepped on or dropped etc. Compact Flash cards are tougher because they have a metal shell and few thin parts, though they can potentially have issues with bent contact pins if inserted incorrectly. It appears that their use is gradually being phased out, even from the professional grade cameras amongst which they used to be ubiquitous – this is not a bad thing since they’re quite a bit more expensive than SD for a given spec and capacity, and they’re also physically larger. Regardless of card type and the fact that none of them have any moving parts, keep cards in some sort of well-padded or hard case to prevent damage; gently cleaning the contacts (SD) occasionally is also an good idea to ensure that the electrical contacts are in good condition. Needless to say, keep them away from liquids and extreme temperatures.

Working with multiple cards
These days, it is highly unlikely that a photographer has a single memory card; if you’re like me and have been shooting for some time, you’ll probably tend to buy larger cards every time you upgrade cameras – partially because the camera upgrade cycle tends to roughly correspond to the amount of time between significant speed and capacity improvements, and partially because you’re simply going to need additional space for the larger files. Eventually, you’ll have a small pile of cards of various capacities; some of them so small (16MB, anybody?) that they’re effectively useless. I just keep the up-to-date cards in my rotation; these are kept in a wallet with individual transparent pockets for the cards to keep them protected and in the same place for easy location. Each camera has its own primary card which always lives there, too. The challenge comes on shoots where you might use multiple cards; my system is that if they’re label-up, they’re fresh and shootable; if they’re label-down, then they’re full and should be downloaded and backed up immediately. Lately though, as my workload increases, I’ve tended to keep the originals on the cards for much longer than I probably should – at least until the job is delivered and the files fully backed up, just in case I happen to need to recover a lost file or something of that nature.

Hopefully, this little article will minimize the risk of something untoward happening to your files! MT


Enter the 2013 Maybank Photo Awards here – there’s US$35,000 worth of prizes up for grabs, it’s open to all ASEAN residents, and I’m the head judge! Entries close 31 October 2013.


Visit the Teaching Store to up your photographic game – including workshop and Photoshop Workflow videos and the customized Email School of Photography; or go mobile with the Photography Compendium for iPad. You can also get your gear from B&H and Amazon. Prices are the same as normal, however a small portion of your purchase value is referred back to me. Thanks!

Don’t forget to like us on Facebook and join the reader Flickr group!


Images and content copyright Ming Thein | 2012 onwards. All rights reserved


  1. Hi MT! Do you have any thoughts on microSD vs full-sized SD cards? Performance and price are becoming more and more comparable between the two (at least from the technical specs published).

    • Oddly enough, no – I haven’t given it much thought as I’ve only got a couple of rarely-used devices that used microSD. I guess if you want flexibility, then go with the smaller one (no idea if the adaptors are UHS-II compatible though) – if you want robustness, go larger. But there probably isn’t any point in buying the smaller ones if you have no devices that use them…

  2. Hi Ming, excellent points! I’ve written a best practice article on memory cards and image safety here: It goes into a bit more detail and touches on some related topics also.

  3. Oddly enough I have just in the last week or so had a problem with compatibility of an SD card with my M8. I cant recall whether or not I knew this particular camera was picky about which cards were used but if I had known, then all I can say is I had then forgotten about it. Until I suddenly started getting random “hangs” during writing to memory, not often but often enough to be worrying. I figured I would assume it was the card not the specific camera (at least as my first option) and swapped it out to another card.So far so good. At about the same time I read this thread and it reminded me about M8 and its quirk in this department. I then went searching on the internet only to find heaps of threads on the same subject by people who had similar issues to me. OK….teach me to read my user manual.

  4. Dear Ming,
    the one I was puzzled with is your ‘protect – delete all’ method. I’d never ever use a delete all command unless after having downloaded and backed up a card after a shoot. What if you forgot to protect one image? Why bother going thrug individual protection in camera? Easier (for me at least) to never delete or, in a pinch (lack of space on a card—in itself, an accident, these days), delete clearly ‘wrong’ images one by one (pressed the shutter accidentaly type stuff). There plenty of time to delete back home…
    As for the rest, I agree with comments: the good old negatives are accessible forever, digital files, who knows how we’ll read them in twenty years (try and read a floppy disk…). Printing the best images is the best way to keep those images accessible to friends and family at all times present and future.

    • Because it allows for very easy curation in camera and faster workflow – the benefits aren’t obvious until you start having to deal with thousands of images per shoot. I don’t recommend it for people who are not accustomed; I’ve been doing this as long as I can remember so it’s reflex. Haven’t accidentally deleted anything yet.

      Printing works, but storage/ display becomes a serious issue.

  5. Iskabibble says:

    The thoughts posted here are the simple ones, the easiest part of backing up. The huge difficulty is once you have your files home on their resting storage media (HDD, DVD, etc). How to then protect them from data rot? Data sitting on both magnetic and optical media can degrade just with time passing. DVD’s that I have burned (on Taiyo Yuden archival quality ones) have read issues after less than 7 years. HDD’s are even more riskier.

    Backing up drives and DVD’s in this case is NOT the solution because you copy along the errors onto the new backup. I’m 35 years old and am making family pictures I want to hand down to my children 35 years forward. They will hopefully do the same 70 years forward.

    How to insure data integrity for such a length of time?

    • It’s the simple stuff that most often gets overlooked.

      As for backups – well, the more the merrier since there is no real proof way of ensuring things will be working 10, 20 years from now. Multiple types of media from the get go is probably the best strategy (and the one I use, now a mix of HDDs, SSDs, prints and my own server).

      • Iskabibble says:

        That’s just it. There is NO way to know the integrity of your digital files. This by far is the biggest issue I have with digital photography. The print from a digital file is about as good as it gets it seems, and even that is not great as ink prints have questionable longevity. I have b & w negatives from my parents and grand parents life from 50+ years ago. I can print them as well as the day that they were made back then. I see no way that digital can duplicate this level of longevity.

        • I agree in terms of viewability. There is one advantage, though: if your negative is toast or badly stored or burned in fire etc – there are no more exact duplicates. At least we have a higher chance of survival with multiple locations with digital.

          The question you pose is a much greater one of permanence of anything: there’s simply no way to guarantee it. The only constant is change.

          • Iskabibble says:

            True, you can increase your odds with digital by creating more than one version but it is a significant management problem to keep track of all these versions as well as keep them fully independent of each other. You just cant dupe your main drive and then dupe the dupe, etc. Doing that makes certain that any corrupt data will eventually overwrite the good. Very few people are up to this level of management. Hopefully industry will find a solution to this extremely complex problem. Imagine 50 years from now retrieving files made from 5 years ago. Are people thinking this far ahead? Or are we just assuming that our files are not worth anything that far into the future?

  6. Speaking of chimping in the field, maybe I was using it wrong, but the GR in RAW-only mode had pretty low-res preview images so I couldn’t judge focus on it. Sony and Olympus do a much better job of this. It is amazing how far displays have come though: my 40D’s display makes every image look terrible and the original Canon Digital Elph’s postage stamp-sized display is hilarious.

    • Ah yes. I have mine set to RAW+JPEG basic for this reason.

      • Thanks. That makes sense, but I didn’t get a chance to try it out before returning the camera. It is a very lovely camera, especially now with the new firmware update.

  7. 8 characters, and not 8 bits for the DCIM file names.

  8. cafelebrocq says:

    Reblogged this on Cafelebrocq Home and commented:
    It’s always good to be reminded of the basics

  9. A few of my memory cards have been through the washing machine (I had a habit of popping them in my shirt pocket) – they came through the process okay.
    The biggest memory card I use is 2Gb – which I seldom fill on a trip.
    Your points are well made – it’s easy to forget the small things – ‘for want of a nail the battle was lost’ and all that!

    • They’re hardier than they look, but not as hardy as we might think or want…and far too easy to lose. Especially those Micro SD things.

  10. I am wondering whether you backup your photos in cloud storage just in case the worst case scenario such as fire, flood or earthquake.

    • No, I don’t trust remote servers because I have no idea who else has access to my data. I carry a portable drive with my critical images though

  11. Always welcome, those rules. Most of seasoned photographer know that, but lazyness rules also… ;-). And each “newbie” finds benefit to read them. A couple of times.And a couple of times more. I only had a card bugging once..

    That’s why I can testify: NEVER write on a memory card through the computer if you want to reuse it in a body and EVER format it inbody.

    • I’ve had some serious card issues before – to the extent of crashing computers into which they were inserted – so yes, this is sensible advice. Especially with the Leicas, which are extremely sensitive to card type and speed…

  12. Tom Liles says:

    I’ve never been stung with lost data, but I’m sure that just means the day it happens is one trip out with the camera closer. But yeah, only ever have one card. That’s because all my cameras, save one, only have a single SD slot. The D3 has two CF slots, of course. I’m only using one. But that will change as I treat myself to a faster CF card next week. I’d just bought the cheapest one to start: figuring that since my D3 was from 2007, the cheapest card in 2013 would be close to what a great card would’ve been back then. I don’t think that’s true—for storage capacity, perhaps yes. But 60Mb/s and up was what it liked then, and they are still the expensive option now. Anyway, I’d never had the photographic occasion to use or need that write speed before, but have had recently and would like to see what the difference is. I’ll probably get two high speed CFs and make use of the two card slots for all the reasons you write above, MT. If we write off the initial outlay for the card: there is no demerit to doubling up.

    I do wonder though: what, then, was the Nikon logic behind the mixed slots on the D4?

    [It couldn’t have been an experiment. Because who experiments with THAT aspect of the pro flagship?]

    • I recently got some 80mb/s SD cards since I’ve now got more devices (and not enough spare cards) – there’s a significant speed difference in the newer cameras. Where I had to wait for the OM-D to write files before reviewing – and was very conscious of the wait – it’s now nearly instant. That extra few dollars is actually quite worth it…

      D4: it made sense since XQD is theoretically a lot faster than CF; the problem is, it’s a proprietary Sony format, and those have the habit of going the way of the dodo. I also suspect it might have been a forced commercial decision that had something to do with sensor technology sharing.

    • We may see more XQD support in the future as 4K becomes more popular, though if Sony is the main company pushing these, then they may not be around long. If XQD disappears, then a used D4 will really plummet in value. So far not a big issue with me. While the D4 is a nicely made camera, I am quite happy to stick with a D3. Recently added the DK-17M viewfinder, and manual focusing is vastly improved over the standard set-up. On the CF cards, I run SanDisk 8 GB for reliability, though I run them sequentially instead of parallel. At my usual shooting pace, especially when waiting for lights to recycle, I never run into writing speed issues.

      • Tom Liles says:

        I got some 8Gb Buffalo thing—it was the second cheapest CF on the shelf, the cheapest was a 4Gb version, same maker same spec. I’ve never come close to filling 8Gb on an outing. On my most frenzied day I think I made 4 and a bit Gb. I occasionally photograph clothing products for my colleagues, not very well but it serves their purpose (and they have 0 budget so couldn’t get a proper craftsman to do it), those shoots eat a lot of Mbs, but those days I have the D3 tethered to LR and it’s all going directly into a computer anyway.

        Just recently though, I’ve had occasion to shoot dynamic “in action” — usually outside somewhere — photos for another colleague’s blog and while I’m not a rapid fire shooter and it isn’t a problem, I would like to test out what difference there is, if any, using a fast card and venturing onto the “CH” drive mode (never used yet; I barely even touch CL).

        Can I add a personal note, Gordon and Ming. Unrelated, my speciality. I printed my first ever photo last week. An optical print of a b&w negative (400 TriX), a photo I took at a night baseball game. Nothing special, just an image I liked. It’s not that perfectly exposed and I doubt the printer did any dodging and burning for me… but I seriously stood back and couldn’t help beaming at it. Really really happy to see something I did, there for real in my hands. The photo suddenly felt more important. They have an RGB “rambda” (?) printer in that print shop, so next challenge is to try a digital one. I might even frame one and hang it on a wall. We’ll see though. That first print is pinned to my wall at work; be interesting to see how long it takes me to get fed up with it… I think this must be a factor in photographic prints; do you just take a punt on which ones you think will have legs, Ming and Gordon? Or do you get a feel for what will look good on paper and what will last?

        • They have an RGB “rambda” (?) printer in that print shop, so next challenge is to try a digital one .
          Lambda. It’s a type of machine.

          …do you just take a punt on which ones you think will have legs, Ming and Gordon? Or do you get a feel for what will look good on paper and what will last?
          Experience, and a lot of testing – the duds which of course I never show. I prefer digital printing because of the level of control and repeatability. But what am I saying…you’ve seen my prints.

          • Tom Liles says:

            Yeah, “lambda” right? As in the greek letter? I’d asked the guy in the print shop and when he told me about the color print options he proudly spoke about their “rambda” machine with parts shipped from Italy; being a beginner I asked what that was and he gave me the spiel… I could only understand about half of it, so I asked him how he spelt “rambda” [so I could check on google later; still hadn’t when I wrote this], the western spelling, and he said r-a-m-u-d-a 🙂

            Good old Japanese phonetics. I put in the silent ‘b’ and should’ve guessed it was just “Lambda.”

            But thanks Ming and Gordon. I am going to try a Lambda print next. When I asked him if they had an accessible shop profile [downloadable, etc] to soft-proof with, he just said “oh yes, it’s AdobeRGB.” So, OK. AdobeRGB. He was a bit frowny when I pushed him on it—he was like, well, I don’t know why you’re so bothered, you’re not going to get 1:1 proofs on screen mate, you have to print a few and get a feel for it. Sure, that’s great for the print shop!
            [And smarts a little from the guy who can’t spell the name of his own shop’s machine; but they are a well respected shop]

            • Actually, AdobeRGB is fine to run on that. I’ve tried ProPhotoRGB a few times, but you should tell them if you are going to try it. If you are not completely sure, then do a small size test print, adjust on your computer, then bring another file for the larger print.

              Lambda uses Red, Green, and Blue laser to expose chemical photographic paper (C-print). Low quality is 200ppi and high quality is 400ppi. Should be using a 50″ roll of paper, unless it is a different machine. This is very different than commercial offset presses.

              • Tom Liles says:

                Thanks Gordon, will be trying a lambda print week after next. His prices weren’t too far off regular digital printing rates actually: for an 8×10, 1400JPY regular digital print, 1700JPY for a ramuda 🙂

                I’ll try both—I’m sure the difference will be perceptible. Can’t wait to have a looksee; and choose a candidate image for the run!
                [the hardest part]

        • Most of my commercial work is intended for print somewhere. Sometimes as large as trade show graphics, or restaurant display. Sometimes as small as 1/4 page in a magazine. It’s not the same as a photographic print, of which I don’t really have that many. Lambda and LightJet are quite good processes, resulting in a chemical photographic print, much like the old processes. I don’t think too many labs are left who do true enlarger prints; it’s no longer cost effective.

          In my advertising related work, most of the images are not intended to have a very long life, because the life of a campaign, or even a trade show, has a limited time the companies want to use them. This need for new images is what can keep me busy. The usual process is to go through the brief and/or storyboards, then figure out the best approach to hit the targets the clients want to reach. Often there will be some extra time, so I may throw in a few of my own ideas and try some approaches. The important thing is to capture what your clients want, and then if you can throw in some other ideas, go that extra step. Many ad agencies and companies go through a long process to figure out what they want, then they bring in a photographer. There is little point in trying to second guess the ideas they have developed. Granted there are times when I get an assignment, and the clients are not really sure what they want, and those are the times when I need to dig really deep in my experiences to pull out some great ideas. As Ming points out, experience and practice are the keys.

          • A lot of the time – in this part of the world at least – we can bring something extra; a sort of coherence if you will – if we get input on the whole creative process…

      • I’ve hit an interesting limit with cards – running 32GBs with spares for each of six cameras, but I have nowhere near that much hard drive space free; it’s almost as though the increasingly larger files force me to be much more selective…

  13. Thanks, Ming! Always good to check basics.

%d bloggers like this: