CF/SD/MS card testing

koffiejunkie

Executive Member
Joined
Aug 23, 2004
Messages
9,754
Reaction score
998
Location
Australia
Question for you guys. Consider the following cycle:

1. Insert blank card.
2. Shoot until card is full.
3. Import all images to your computer.
4. Erase or format the card.

How many times do you need to repeat the above cycle before you are confident that the card is not defective and will not fail shortly after you start using it?
 
I was under the impression that flash media had a finite number of write/erase cycles so the more you use it the more likely it is to fail.
 
It has, but that's of academic interest in the context of my question. It's in the thousands. We talked about testing cards before, and you mentioned that Spinrite is not recommended.

I'm trying to come up with a way to test a card in as close to normal usage conditions as possible, without incurring unnecessary load on it. Hence my question. How many times times do you use a brand new card before you're confident the card isn't a dud?
 
It has, but that's of academic interest in the context of my question. It's in the thousands. We talked about testing cards before, and you mentioned that Spinrite is not recommended.

I'm trying to come up with a way to test a card in as close to normal usage conditions as possible, without incurring unnecessary load on it. Hence my question. How many times times do you use a brand new card before you're confident the card isn't a dud?
If it reads and writes the first time I'm happy with it. It's got no moving parts so any failure should be present from the start, right?
 
It has, but that's of academic interest in the context of my question. It's in the thousands. We talked about testing cards before, and you mentioned that Spinrite is not recommended.

I'm trying to come up with a way to test a card in as close to normal usage conditions as possible, without incurring unnecessary load on it. Hence my question. How many times times do you use a brand new card before you're confident the card isn't a dud?
I would:
Create a file full of random data on the HDD. Get the md5 of that.
Write file to card.
Read file off card and then compare with the md5.

Loop that a 100 times with new random data every time. If the read/write of the card is error free then it should be OK for images too.

But in practice I wouldn't bother.
 
I would:
Create a file full of random data on the HDD. Get the md5 of that.
Write file to card.
Read file off card and then compare with the md5.

Loop that a 100 times with new random data every time. If the read/write of the card is error free then it should be OK for images too.

But in practice I wouldn't bother.
Same here but then I only shoot with 1 and 2gb cards so I dont lose everything if/when one fails.
 
Ok, so are you guys happy to stick a new card in and go shoot a gig? Or do you go out looking for some random stuff to shoot to fill up the card to know it's OK?
 
I would:
Create a file full of random data on the HDD. Get the md5 of that.
Write file to card.

The problem with that is the file would be limited to4GB (FAT32). And (I'm not sure if this really matters, but I suspect it might). You're testing one write operation. What I had in mind is taking a bunch of images from my library, shot with the camera I'll be using it in (RAW or JPEG, which ever you shoot), write that to the camera.

Read file off card and then compare with the md5.

you can check the md5 directly on the card - it's already a full read.
 
i still have my 128MB card that i bought in 2003
i now use it for transferring music to the hdd video machine
lol - it still works
 
Hehe, my mom's still using the 64MB card she got with her 4MP Cybershot. I'm sending her my old 512MB along with my old 4MP Cybershot (hers is broken - same model).
 
Hehe, my mom's still using the 64MB card she got with her 4MP Cybershot. I'm sending her my old 512MB along with my old 4MP Cybershot (hers is broken - same model).
I have a Sony F707, and the max size card it can handle is 128MB :eek:.

As far as testing is concerned, I would say if you filled the card once successfully you're ok.
 
As far as testing is concerned, I would say if you filled the card once successfully you're ok.
+1
I'd test it once. Fill it with data and then defrag it manually. If there are any bad sectors you'll see it immediately.

No point in stress testing, you're just reducing the lifetime.
 
Top
Sign up to the MyBroadband newsletter
X