Does SD card speed matter?

[holmework]

Having bought a Panasonic LX3, its time to buy a bigger SD card. Out of habit I try to buy the fastest (within reason), but I am finding it hard to find speed comparisons online. So I wondered if that is because the world has concluded that speed doesn't matter any more?

If I shoot RAW+jpeg, its saving 15-ish MB. On the fastest card that would be just over 1 second, and on a slow card maybe 2-4. I would think thats a worthwhile time difference. But unless I shoot continuous, or bracketed, 1-2 seconds is within the time it takes to look at the preview. Is this what other people have concluded?

Does anyone know of any good speed-comparison sites?
I have seen:
http://www.robgalbraith.com/bins/cam...?cid=6007-9424
and
http://techgage.com/article/sd_card_roundup/2
and
http://www.tomshardware.com/reviews/...card,2143.html
They all seem to give different results, and are varying-degrees of out-of-date.

[Tooslow]

h, all I can say is that I've just bought a Sandisk Extreme III (from Mymemory) rated at 30MB/s. "Up to" should no doubt be read into that. It is only marginally "faster" than a Sandisk Ultra. No speed marked on it but I suspect "up to" 10MB/sec.

If I want the camera to zip along really quickly the best thing I can do is turn off D lighting at which point it will shoot continuously until I get bored. With D lighting on the Ultra card will take 5 shots and then take 15 seconds to clear the buffer, the Extreme, 6 shots and 10 seconds.

I'm deeply suspicious of any manufacturer that describes speed as "times". Times what? How do I know that Sandisk and Lexar are both 50 times faster than the same thing? Or "Class". Is higher better or worse? And what does it mean? Honest Mb/sec please chaps. Then we can compare.

So, turn off the frills on your camera if you want to really speed it up. If you do spend money on a faster card, don't expect miracles and I do recommend mymemory.co.uk, I have no connection with them other than a recent purchase.

JH

[El_Sid]

h, all I can say is that I've just bought a Sandisk Extreme III (from Mymemory) rated at 30MB/s. "Up to" should no doubt be read into that. It is only marginally "faster" than a Sandisk Ultra. No speed marked on it but I suspect "up to" 10MB/sec.

Have you checked the write speed of your camera? If the maximum write speed of the camera is less than that of the card then the maximum performance of the combo will be that of the camera - which may itself be only marginally faster than the Ultra card...

If I want the camera to zip along really quickly the best thing I can do is turn off D lighting at which point it will shoot continuously until I get bored. With D lighting on the Ultra card will take 5 shots and then take 15 seconds to clear the buffer, the Extreme, 6 shots and 10 seconds.

Additional in-camera processing (D-lighting, High ISO NR etc) does tend slow down the camera. My 40D runs fastest in JPEG with highlight tone priority (or whatever it's called) turned off. Switch to HTO on and RAW and it's way slower...

Your figures work out at an extra 16% capacity and 33% faster writing - which doesn't seem unreasonable. I also note that the SanDisk site gives 30MB/s as the speed for the Ultra card too...

[holmework]

According to wikipedia (yes, I know!), there is a fixed speed that they are all multiples of:
http://en.wikipedia.org/wiki/Secure_Digital#Speeds
But I would imagine they are achieved under 'test' conditions.

Interesting what you say about D-lighting. I see if it makes any difference.

[IvorETower]

Nikon themselves state that having D-lighting on significantly slows down the shooting rate; also as mentioned above, the write speed that the camera is capable of achieving will affect the attainable frames-per-second rate

[Nanette]

I always shoot in RAW with no JPEG on the 5D MK II, i have not tried a slower rated card as i think the RAW format is slow enough with a faster rated card. I use lightroom to preview my raw files so dont need a jpeg.