Memory cards

[Mike Buckley]

I have always used the card reader built into the computer and have never had an issue. The last two computers included four types of readers, though I've never used anything other than SD cards.

[AlwaysOnAuto]

The idea of just "tossing" CF cards into a camera bag is ludicrous. Not so much because the cards may be physically damaged but because they could easily be lost. I use a small fifteen year old OPTECH filter wallet which allows me to store and keep track of the cards.

When I go on a small shoot, I always like to have an extra CF Card and battery with me so I place the CF Card in the watch or change pocket of my jeans and carry the battery on my belt using the case for a no longer used Flip Phone.

I just need to remember to take the card out of the pocket prior to washing the jeans

Just remember to run it through the dryer if you do that Richard and you'll be fine!

[lunaticitizen]

I would have to disagree with you as I have had issues #4 - cheap card reader destroying data on my card. I now only use a Lexar reader.

I have had "no name" cards fail in the field, so I also agree with #8. I now only use Lexar and San Disk cards. As Lexar is exiting the business, I am not going to buy their cards until we see what happens when a new owner (if there is one) takes over. The only time I have had a failure with a high end card is when I used it with a cheap reader (i.e. #4). I was able to recover most of the files, but the cheap (and cute) reader my daughter gave me went into the trash right away. No permanent damage to the card, so far as I can tell.

Okay, I probably exaggerated a bit on my previous post. I mean, it was really nonsense up until the third one I didn't really pay attention to the rest of the article.

About #4, I have never bought a no-name card reader either, but a cheap one I've had for years served me without fault, until I lost it a few months ago. Yes, this advice is sound.

#8 is similar to #4; I also avoid no-name products yet I don't equate cheap with lack of quality. I've had generic flash memory products from Toshiba, Sony, or Buffalo and I am satisfied with them. I couldn't care less about the crème de la crème from Lexar or Sandisk. No, I'm not a sport shooter so I don't need extreme speed.

[Manfred M]

#8 is similar to #4; I also avoid no-name products yet I don't equate cheap with lack of quality. I've had generic flash memory products from Toshiba, Sony, or Buffalo and I am satisfied with them. I couldn't care less about the crème de la crème from Lexar or Sandisk. No, I'm not a sport shooter so I don't need extreme speed.

I've had branded cards from Verbatim and Patriot fail; one quite recently. Neither manufacture their own cards. San Disk and Lexar do and also make their own memory..

[rpcrowe]

I have always used the card reader built into the computer and have never had an issue. The last two computers included four types of readers, though I've never used anything other than SD cards.

Mike,

I really don't know much about SD Cards, in fact I don't even know the speeds of the SD Cards that I own.

However, with relatively large files such as those from my 7D2 and my 5D2, I have noticed a great difference in the write speed between faster cards and slower CF cards. However, when shooting with a 30D or 40D camera, I did not notice any difference in the time it took for my buffer to clear. It did not matter whether I was shooting slow or fast cards. When I traveled to China, I shot with 4GB Lexar 300x speed cards which were UDMA compatible.

Neither camera could take advantage of UDMA technology but, when downloading the files using a UDMA capable card reader (at that time it was a Lexar USB 2.0) there was a significant time difference in downloading the files between UDMA cards and non-UDMA cards. I have not made any comparisons between download speed of the USB 2.0 and USB 3.0 readers.

I have notice one thing about readers in general. Some readers, like the Lexar 2.0 and 3.0 models have a longer slot into which you slide the CF card. Others have a very short slot before the card comes in contact with the pins. IMO, the longer slot is safer because there is less chance of the card being inserted at an off angle, thus damaging the pins.

All of my present first-line cards are at least 300x, 400x or 800x and are all UDMA capable. I have some back-up cards left over which are usually 2G and significantly slower and not UDMA capable. They still work fine, just don't have the speed of the more modern cards.

[Mike Buckley]

Thanks for the info, Richard, but the write speed of the cards is of no concern to me because of the type of photography that I do. Similarly, the speed of transferring from the memory card to the computer is of no concern to me because I am always relaxing or getting something else done while the image files are being transferred. Indeed, while they are being transferred, initial cataloging is also being done automatically, which slows up the speed even more.

[rpcrowe]

I do shoot a bit of action and, although I don't shotgun my images by keeping my finger pressed on the shutter button, I do end up with a lot of images. It's nice for the card to clear the 7D and 7D2 buffers quickly. I haven't experimented with my 5D2 because I generally use that camera for portrait work and don't need the fast speeds.

When traveling, it is not unusual for me to shoot at least hundred GB of RAW images per day and I download these in the evening. Since I am an old man, I want to get to bed and the sooner my images download, the sooner I can turn in

[Black Pearl]

I delete images in camera - have done for as long as I've used digital camera - I typically use a card reader, I pull the cards out all the time without ejecting them, rarely if ever format a card when I put it in a camera unless I need to clear it and have never had any issues I'd associate with doing so. The 'advice' in the link crops up all the time but I have yet to see absolute out-and-out proof it is better.

Ps. In all the camera stores I have worked you would pick up a card and find dozens of folders on them from dozens of different camera they had been in and they all worked flawlessly.

[rpcrowe]

I do however, try to remember to use the computer's "safely eject" command when I remove the card reader. I have not had any problems with this computer but, did occasionally have a problem with the previous computer I was using which was a Gateway. My tech-guy said that I should always use the safely eject command prior to ejecting media. Since I have had no problems with the Dell computer, I have been fairly lax and have noticed no problems...

[NorthernFocus]

No doubt the things that are pointed out in the article are theoretically accurate. But each individual must decide whether they are relevant from a practical standpoint based on our own operating context. A context which may change with time or with a given shooting situation.

A good portion of my professional career involved making risk based decisions (then again most decisions that people make every day are managing risk though few think of it that way). Risk is simply probability versus consequence in a given context. That said, two of the most common drivers that people use in making risk based decisions are (likely) the least relevant.

- xyz has never happened to me therefore it never will. And,

- xyz happened to me once therefore I will NEVER let that happen again.

We humans are fickle

In the mean time, it's perfect subject matter for a forum discussion. There's no "correct" answer

[Shadowman]

I will also say that not rushing to format a card after transferring has sort of been imbedded in my workflow, one time during a lengthy shooting session I transferred images to the computer and formatted afterwards; I didn't realize until later that my camera had started another folder and I totally missed it during the transfer. Luckily I only lost a few dozen images that time. Now I check every folder created, for some reason I have four folders on my one memory card even though only two are ever used. I'll get around to deleting the extras one day.

[Mike Buckley]

- xyz has never happened to me therefore it never will. And,

- xyz happened to me once therefore I will NEVER let that happen again.

I often think of a possibility that is in between those two scenarios: xyz has never happened to me, and though I know it could happen to me, I'm willing to accept that risk in return for not having to prevent it from happening. As an example, I never back up files on a memory card while traveling. The requirements for doing so just aren't worth it to me. I realize that I may change my mind about that once I get stung, but I can't predict the future about how I will likely react to a situation I've never experienced.

[pnodrog]

I will also say that not rushing to format a card after transferring has sort of been imbedded in my workflow, one time during a lengthy shooting session I transferred images to the computer and formatted afterwards; I didn't realize until later that my camera had started another folder and I totally missed it during the transfer. Luckily I only lost a few dozen images that time. Now I check every folder created, for some reason I have four folders on my one memory card even though only two are ever used. I'll get around to deleting the extras one day.

I am the same. I do not format the card until I have confirmed that the photographs are safely on my computer and have been backed up. For important sets of photographs I replace the card after transferring files from the camera. It will sit in my bag as an emergency backup and is not formatted again until I have filled the replacement card.

Downloads to Lightroom certainly work quicker if there are no previously downloaded photographs on the card so I tend to format the card as soon as I am happy that the photographs are securely backed up.

My 32GB CF card gives me about 1100 NEF images and unless I am traveling I seldom have more than about 200 photographs on it.

[DanK]

No doubt the things that are pointed out in the article are theoretically accurate. But each individual must decide whether they are relevant from a practical standpoint based on our own operating context....

That said, two of the most common drivers that people use in making risk based decisions are (likely) the least relevant.

- xyz has never happened to me therefore it never will. And,

- xyz happened to me once therefore I will NEVER let that happen again.

We humans are fickle

In the mean time, it's perfect subject matter for a forum discussion. There's no "correct" answer

I certainly won't argue that people evaluate risk accurately or respond to it rationally in daily life. There is decades of research showing the people generally don't. However, I don't think this thread is only about the difficulties in assessing risk. There are parts of this that have correct answers, or at least correct data, and I am not convinced that the author of the article provided them. For example, I would like to know whether deleting in camera actually DOES mess up the FAT and, if it does, whether it does so at any non-negligible rate. If it does, then I would want some data on frequency--or risk.

[xpatUSA]

I certainly won't argue that people evaluate risk accurately or respond to it rationally in daily life. There is decades of research showing the people generally don't. However, I don't think this thread is only about the difficulties in assessing risk. There are parts of this that have correct answers, or at least correct data, and I am not convinced that the author of the article provided them. For example, I would like to know whether deleting in camera actually DOES mess up the FAT and, if it does, whether it does so at any non-negligible rate. If it does, then I would want some data on frequency--or risk.

As a consulting engineer at Rolls-Royce in Mount Vernon, Ohio, one of my tasks was to calculated "risk mitigation" for various bits and pieces.

I can assure this forum that, just because something bad hasn't happened yet, it certainly will with some probability factor, never ever stated here, of course.

So, statements like "I always do this and nothing bad's happened yet" carry no weight with me, sorry.

[rpcrowe]

I guess that this might be a place to mention this...

Several times lately, my RAW files as they downloaded have been separated into folders of different dates. One day apart. I checked to see if the internal clock/calendar of my camera was set correctly and it is. I have no idea why the folders have been separated like this.

[pnodrog]

I certainly won't argue that people evaluate risk accurately or respond to it rationally in daily life. There is decades of research showing the people generally don't. However, I don't think this thread is only about the difficulties in assessing risk. There are parts of this that have correct answers, or at least correct data, and I am not convinced that the author of the article provided them. For example, I would like to know whether deleting in camera actually DOES mess up the FAT and, if it does, whether it does so at any non-negligible rate. If it does, then I would want some data on frequency--or risk.

This mostly from my fading memory....

Deleting a file simply writes a Hex E5 to the leading character of the file name in it's entry in the directory. This entry also points at the starting FAT entry for the file. I don't think that the FAT file is changed at all at this point. When the card/disk is nearly full the operating system rewrites a deleted file entry in the directory, changes it's name, writes the new file using the existing FAT entries. If the file is smaller than the previous file it should clear all linked FAT entries no longer used and place an end of file marker Hex FFFF in the last FAT entry it is using. At this point the previous file has either completely overwritten or blocks/sectors that have not been overwritten will be marked as available (Hex 0000) in the FAT file. If the file is longer than the previous file it uses all the FAT links of the previous file and then looks for empty FAT entry's that indicate unused blocks it can continue writing the file to and adding a new link or EOF (end of file) marker to the FAT entry. If the link from one FAT entry does not link correctly to the next FAT entry for the file the rest of that file is lost and in most cases the file is corrupt. Even worse than putting a book on a library shelf and loosing it's Dewey number.

It is system that should not go wrong if all the software does it's thing and there are no power failures or other glitches while doing it.

Actually memory cards have their own processor that is optimised to extend the cards life and try to limit rewrites as much as possible so to the operating system they appear to be behaving like a normal FAT memory devise but maybe using all sorts of other methods to save and retrieve data.

FAT failures will be most likely to occur (if at all) once the memory card has nearly reached it's full capacity and starts using the space available from deleted files.

Overall the risk of failure of a FAT is very low but it is subject to variables, such as RFI, software, power failure, temperature, file sizes, operations since last format, etc etc that in some situations the risk would be far greater than the statistics may indicate. A bit like driving statistics, a few drivers have far more accidents than the statistics would indicate and others have none.

My personal approach is to delete as little as possible and format the card in camera as soon as I have confirmed I have the previous images saved and backed up.

P.S. The structure of the FAT system depends on which standard is being used but the strategies in using it are dependent on the operating system and drivers etc.

[NorthernFocus]

This all from my fading memory....

Deleting a file simply writes a Hex E5 to the leading character of the file name in it's entry in the directory. This entry also points at the starting FAT entry for the file. I don't think that the FAT file is changed at all at this point. When the card/disk is nearly full the operating system rewrites a deleted file entry in the directory, changes it's name, writes the new file using the existing FAT entries. If the file is smaller than the previous file it should clear all linked FAT entries no longer used and place an end of file marker Hex FFFF in the last FAT entry it is using. At this point the previous file has either completely overwritten or blocks/sectors that have not been overwritten will be marked as available (Hex 0000) in the FAT file. If the file is longer than the previous file it uses all the FAT links of the previous file and then looks for empty FAT entry's that indicate unused blocks it can continue writing the file to and adding a new link or EOF (end of file) marker to the FAT entry. If the link from one FAT entry does not link correctly to the next FAT entry for the file the rest of that file is lost and in most cases the file is corrupt. Even worse than putting a book on a library shelf and loosing it's Dewey number.

It is system that should not go wrong if all the software does it's thing and there are no power failures or other glitches while doing it.

Actually memory cards have there own processor that is optimised to extend the cards life and try to limit rewrites as much as possible so to the operating system they appear to be behaving like a normal FAT memory devise but maybe using all sorts of other methods to save and retrieve data.

FAT failures will be most likely to occur (if at all) once the memory card has nearly reached it's full capacity and starts using the space available from deleted files.

Overall the risk of failure of a FAT is very low but it is so subject to variables, such as RFI, software, power failure, temperature, file sizes, operations since last format, etc etc that in some situations the risk would be far greater than the statistics may indicate. A bit like driving statistics, a few drivers have far more accidents than the statistics would indicate and others have none.

My personal approach is to delete as little as possible and format the card in camera as soon as I have confirmed I have the previous images saved and backed up.

P.S. The structure of the FAT system depends on which standard is being used but the strategies in using it are dependent on the operating system.

So which part of your memory is failing?

[pnodrog]

So which part of your memory is failing?

I am glad I edited it from "all from" to "mostly from" before I saw your comment. There were a couple of things I did have to check...

[dearjoyce]

Some part of the content requires common sense, but it really is worth reading. Remind me again who delete the photos one by one instead of inserting it on our laptops to filter which is a great shot and which is not? LOL. It's like the do's and don't's, thank you for sharing this good content with us.