SD card backup?

My netbook only has a 10 inch screen so is only a bit bigger than most tablets, especially those with a keyboard and cover.

I leave it either in plain view on the desk, or in my camera bag or suitcase. Camera gear lives in my camera bag if I don't have it with me, or out on the desk.

Hotel safes are easily opened and relocked, so you might as well leave the stuff out.

We don't stay in hotels, we rent cottages, so there are plenty of places to keep the laptops.

Some interesting stuff. I've also been of the opinion that middle of the road capacity cards of around 8gb are probably the safest bet. I've also stopped re-formatting my cards since I read somewhere that it could reduce the cards life? I now just delete the images rather than format. I have to say that I've not had a card, either CF or SD fail yet. Perhaps there's still a place for an independent storage device for those of us that don't need or want to lug a laptop around?

Just deleting the images means the card capacity may gradually reduce. Formatting resets the file structure.

In any event, our cards have outlived their useful capacity long before they have failed. We moved from 1GB to 2, 4 and now 8GB. Next time I replace all the cards, which I do every couple of years, no doubt it will be 16GB.

I lost a backup hard drive a few months back, two years worth of images! Lucky I'd learned that lesson a few years previously and it was only one of two backups.

I've been just deleting the files off my 16Gb SD Cards for years and they still hold 16Gb.

Why would cause the capacity reduce by just deleting all the images?

What would make a larger capacity card more likely to fail than a smaller one?



.

The underlying technology (in this case bit density). Just like a FF 16Mp sensor offers lower noise/higher dynamic range than a 16Mp APS-C sensor, a larger card is likely to be a higher bit density than a smaller one. The smaller the bits, the more likely they are to be corrupted.

Compact Flash cards are likely to be more reliable than SD(HC) cards, partly for this reason, partly because they have a standard IDE interface (which removes an additional 'corruptable' layer.).

John.

There's no evidence that squeezing more transistors/resistors/capacitors/inductors on the same area of silicon, squeezing more pixels on the same area sensor, squeezing more bits on hard drive platters, etc has reduced reliability rates.

If it were the case the electronics industry would have collapsed long ago. SD Cards were only 1MB when they were introduced. You can now get 256GB cards. That's a 256,000 increase in density.

I remember 20Mb Hard drives ( I've still got one somewhere and it wouldn't even hold one average RAW file ). you can now get 4TB HDDs, that's a 200,000 density increase.

And all the while the reliability of electronic components has only got better, not worse. I see no evidence of serial interfaces being less reliable than parallel interfaces either.

IMO it's irrational to be less confident of electronics becuase the components are smaller

There are a couple of other things to consider.

The mechanical act of inserting and removing a card it likely to cause a failure, either in the fabric of the card or the contacts. In my experience I've had a cheap card fall to bits but the 32GB Sandisk ones are pretty solid and waterproof to boot.

The more you insert, move, remove, load etc etc the more likely you are just to lose the damn card!!!

I'll stick to the 2x 32GB and a couple of 16GB ones.
When I upgrade I let the family have the old ones.

When I'm out east I take a lightweight basic laptop which allows use of e-mail & skype as well as being the storage medium for each day's work - it is a lot harder to lose a laptop than an SD card. Well, that is my logic, anyway

The mechanical aspects of the card (or reader/writer) are far more likely to fail than the silicon. The aim of the silicon manufacturer is to keep the silicon as small as possible, not only for economy (less silicon costs less) but also any mechanical stresses are reduced by using a smaller chip (for the same deflection of the card housing a smaller chip deforms less - equals less mechanical stress on the chip).

I've only ever used Sandisk cards (CF or SD[HC]) in my cameras and haven't had one fail yet, my earliest ones (CF) from circa 2000 are still readable. I've had non-Sandisk cards for other uses fail within a week.

The most likely reasons for a card to be corrupted are faulty (or cheap) manufacture, incorrect read/write interfacing from the reader/writer or user error. No matter how well the 'write' software is implemented, there's still likely to be a 'one in ten thousand' (or hundred thou or million...) chance of a mis-write - which is as likely to happen on the first or millionth write of any one card.

John.

Indeed. When I hear of people recommending to flick the card out of a K-5 because they find the normal method between thumb and forefinger awkward I cringe as that's probably the best way to start causing cracked print and other mechanical failures.


... but cards are not more likely to fail the more capacity they hold.

If there's any rationale for using smaller capacity cards, it's not putting your eggs in one basket, so if you break a card from flicking it out o the camera or other mistreatment, lose it, or it fails for other reasons you don't lose the whole shoot.


I use Integral Ultra class 10 16GBcards. None have ever failed

It's the same as not defragmenting a hard drive, the card gets cluttered with odd bits of old file structure. Formatting sets everything up anew, just as formatting a hard drive would.

Some might consider it overkill, some might think it's just good housekeeping. I think the latter, certainly where SD cards are concerned.

For those that subscribe to the "not all in one basket" philosophy this seems a pretty good deal.

http://www.7dayshop.com/sandisk-ultra-sdhc-memory-card-8gb-class-10-uhs-i-30mb-s...