EOS 5D CF-Card interface speed

[horst_s.]

Dear all,

do you have experiences with the CF-Card interface of this new

camera - is it as fast as a 20D or 1D? Or is it even as slow as a

10D? Any suggestions about the data rate (MBs per second)?

The question for me is: Is it worth to invest in fast CF-cards?

As a owner of a 10D I know that if the card interface is poor, even

the fastest card will not win the battle.

 

Rob Galbraith hasn?t updated his CF-database with the figures for

the 5D.

 

Your comments are welcome!

 

Regards

 

Horst

[steve_dunn2]

<p>I'm just guessing here, but I'd be very surprised if it weren't at least the equal of the 20D. DIG!C II is supposedly responsible for the speed improvement from the 10D and 300D to the 20D and 350D, and the 5D uses DIG!C II.</p>

 

<p>Whether it's somewhat faster than a 20D, as the 1D II and 1Ds II are, or not, I wouldn't want to guess.</p>

[horst_s.]

Sorry,

my question was not correct. Want I want to know is, if it makes any difference to use a fast CF or a slower one, for example a Sandisk ultra II vs. a microdrive.

The interface of the 10D is that slow, that it does not make a remarkable difference.

Regards

Horst

[ray robertson]

My experience is a fast CF card makes a practical difference in situations where you are doing a lot of continuous shooting and auto bracketing. The last time I came across the limitations of a slow card (a 4 yr old IBM microdrive) was when I was doing a series of pano shots for stitching purpose and using AEB at the same time. At each position, I was taking a burst of 3 shots and then swivelling the camera quickly to the next position. Because the data was not being written quickly enough, I found myself having to pause a few times for the buffer to clear.

 

You'll also notice a difference if you are in the habit of downloading images directly from the camera.

[steve_dunn2]

<p>Personally, I just about never shoot bursts which are larger than my 20D's buffer, and I don't particularly mind if chimping and downloading are on the slow side, so for my uses, money spent to buy faster cards would be money wasted.</p>

 

<p>But you have a 10D, which writes more slowly, and you may shoot in a different way than I do. The net result is that your 10D is too slow for you. Fair enough. That means that when you upgrade to a new body, you need one which will write more rapidly than the 10D (and I'm certain the 5D will), along with a card which will give higher performance. If you already have a card which is faster than your 10D can write, you should see a boost in performance simply by upgrading the body. If that's not enough of a difference, you'll need faster cards.</p>

 

<p>You ask what the data rate is likely to be. Chances are that it will be at least as high, with any given card, as the 20D, so you can use the 20D section of Galbraith's database for a first estimate. You can also use the 10D section of Galbraith's database to get an idea of how much faster a given card performs in the 20D vs. the 10D, and again, since the 5D is likely to perform as well as the 20D if not somewhat better, that gives you an estimate of the 5D's performance.</p>

 

<p>You also ask if it makes a difference to use a faster card. Yes! The data for the 20D show that quite clearly; the fastest card tested in the 20D performs over 5.5 times faster than the slowest card. The performance of the slowest card is limited by the card; the performance of the fastest card is limited by the body (since the fastest card is 80x but the actual performance tops out at about 40x). The 5D won't be able to make a slow card any faster, so the slowest card will give essentially the same performance in the 5D. It may be able to take better advantage of faster cards, so the spread between fastest and slowest cards in the 5D might be 6x or 7x.</p>

[horst_s.]

Steve, Raymond,

thank for your ideas. I should look for a local dealer with a 5D and then fire till the end of the buffer (using different CFs).

If I could manage that, I will let you know the result.

 

Regards to you

 

Horst

[mark u]

One thing to bear in mind with the 5D is that you have the benefit of a fairly large buffer that will handle 17 RAW shots (and many more - even unlimited - jpegs, depending on quality setting) at 3fps. That means that you are only likely to find yourself being constrained on the basis of the number of shots per minute that you fire, rather than being caught by the shallow buffer of the 10D that gets exhausted in a couple of seconds. Given the larger file sizes, you are probably going to want to consider 2GB or larger cards if you shoot a lot of bursts anyway. Cards of that size don't come in versions that are very slow. Then it's a question of how impatient you are when you feed the card into your card reader to download images.

[gary_j1]

The SANDISK EXTREME III is 20 MB/sec and will keep up with any camera, I would think. I don't have the 5D, but I use these cards in my XT and 20Da. With a life-time waranty you can shoot away and be ready for your next camera too, without worry.

 

If there is a speed problem, at least you'll know it's not the card!<div></div>

[happy_chappy]

I wondered this when I got my 5D. I had cf cards up to 1gb in size and with the increased file sizes of the 5D I guessed I needed a bigger card - but what speed? I opted for an i-Pro 2gb 15mb/s 100x speed.

 

Thinking about it this way, the 5D manages 60 jpegs buffered and about 17 RAW. It isn't very often that you are going to fill the buffer with those capacities. So do you need massive speed to save your images? Possibly not. Where I have noticed the difference is in the image review time. The 5D with a 1gb Kingston the image review time is quicker than the 10D. With the 5D and the i-Pro it is practically instantaneous.

[horst_s.]

Dear all,

 

I managed to find a local dealer where the 5D is on stock (last one).

I did the spped test with an ibm microdrive (1GB) and a sandisk ultra II (2GB).

The microdrive got the buffer empty within 83 sec. For the ultra II it took 53 sec.

So the result is: "Speed matters".

You are right, under normal circumstances there is only a small chance to fill the buffer totally.

By the way. Picke up the 5D for a dream price of 2800 Euros.

 

Regards and thanks for your support

 

Horst