KiboOst wrote:
Is there a mode where 1Dx will write, in a burst, img1 to slot1, img2 to slot2, img3 to slot1, img4 to slot2, etc ? With two fast cards this would be lot longer burst, no ?
I personally find it more interesting how many raw photos you take before the fps takes a dive and how soon the buffer is cleared again, but each to his own.
I recently tried 1D X with the Lexar x1000 32GB and the combination could take around 50 shots in raw @ 12 fps before the speed was slowed down. Good enough for me.
Sven Jeppesen wrote:
Thanks for testing the different CF cards Lars.
I like the 30 seconds test of them. Because that's what I have been reading at Rob Gailbraiths site when he was testing cards for many years. Personally I can't see the point in making a test where you would shoot only few seconds. Then all cards would show nearly the same result !!
Gunzorro wrote:
Thanks Lars! It's nice to have some practical yardstick to gauge these newer cards. It would be nice if we had a summary for which CF/SD cards are most appropriate to which camera bodies, from say 1Ds2 on up through 1DX/5D3/T5i models. I gave you a like on your last post, but it applies to the whole thread.
Dan -- I'm surprised you don't "get" the reason for the testing procedure (30 sec) to reduce time variables that would crop up in tabulating exposures in shorter bursts. While the 30 sec does have some application for action shooters, it is being used to quantify test results, not as propaganda encouraging "spray and pray" technique. Once again, you've turned this into a treatise against gear acquisition and individual testing.
Gunzorro wrote:
Following up on my comment regarding cards vs. models: I've got a 1Ds3 and 5D2 as my two top MP cameras. Not anything fancy by today's standards. I use Transcend 32GB 400X in both (based on Lars recommendation some while back). They seem fine to me for in-camera and uploading via my USB 3.0 cardreader.
Is there anything to be gained by moving to faster cards for a "methodical" shooter like myself, who doesn't shoot in long bursts? I never fill the cards, and could get by with 16GB in about 95% of my use (8GB about 90% of the time). Is there anything practical to be gained for someone in my position by going to a larger or faster card?
I know this is opening up onto a huge subject, and many of us just bumble along in our ruts of knowledge. ...Show more →
I belive you have the perfect cards for your bodies and shooting Jim You will not gain much in buying the fast 1000x cards
twistedlim wrote:
Great tests, thanks for the effort. Sad think is for me I have become lazy and been using the slow sd cards since there is a handy little slot in my macbook. For paid events it will be only CF cards on the 5d3 so this really helps.
I have aways used the sandisk extreme cf cards. Trying others that are supposed to be much faster and always coming back to them. Good to see they are solid on both cameras.
stanj wrote:
Thanks Lars for all the testing and explaining. I did something similar last July when the 1DX came out and also "didn't do it right"
One thing that deserves a mention is that the faster cards (in my case, Lexar 32GB 1000x) will make the camera shoot forever at 4fps or so. For some uses, 12fps is overkill, but something like 4fps is perfectly fine and you simply never run out of buffer. I understand that some people say that if you keep running out of buffer you don't know what you're doing, and they are entitled to their opinion (and who knows maybe they're even right). But different people have different needs and thus the different tests. So again, thanks for yours
I personally find it more interesting how many raw photos you take before the fps takes a dive and how soon the buffer is cleared again, but each to his own.
I recently tried 1D X with the Lexar x1000 32GB and the combination could take around 50 shots in raw @ 12 fps before the speed was slowed down. Good enough for me.
My tests on the Lexar 1000x and Sandisk Extreme Pro (both 32GB) indicate the same thing. There was not a large difference between them.
Lars Johnsson wrote:
I did a test with my 1DX body. Every setting exactly like the older test with my 1D4. And the same 30 second burst. Some surprising results compared to the other body.
ISO 100
1/4000 sec
f/5,6
lens EF 24L
Gitzo tripod RRS head
Hope I am not stepping on anyone's toes here, but I did a calculation based upon Lars' data for the 1Dx.
Assumption: 1Dx buffer is 30 frames. Camera starts writing after the 1st shot and continues writing through the 30 second test. At the end of the test the buffer is full and the camera has written the number of shots Lars recorded less 30 which are still in the buffer.
I then calculated how fast the camera can shoot when the buffer is full.
For example in the case of the Lexar 1000x card, the initial 30 shots filled the buffer. The next 130 (160-30) shots occurred as the buffer was emptied. For every shot taken, one had to be written to the buffer. The buffer was being emptied for the entire 30 seconds of the test, so 130 shots written to the buffer / 30 seconds = 4.3 FPS when the buffer is full.
RichFisher wrote:
My tests on the Lexar 1000x and Sandisk Extreme Pro (both 32GB) indicate the same thing. There was not a large difference between them.
You have to remember that Sandisk Extreme Pro have two versions of that card. One UDMA 6 version and one UDMA 7 version. And there is a rather large difference in speed between those
Lars Johnsson wrote:
You have to remember that Sandisk Extreme Pro have two versions of that card. One UDMA 6 version and one UDMA 7 version. And there is a rather large difference in speed between those
how do you tell the difference between the 2 versions. All my cards say 90 MB/sec and UDMA 6. B&H shows a card which still claims 90 MB/sec but claims UDMA 7. Still 90 MB/sec. In the real world what is the difference?
I personally find it more interesting how many raw photos you take before the fps takes a dive and how soon the buffer is cleared again, but each to his own.
I recently tried 1D X with the Lexar x1000 32GB and the combination could take around 50 shots in raw @ 12 fps before the speed was slowed down. Good enough for me.
stoffer
There are really 2 questions here. Yes hope much can you shoot at max speed but also once the buffer is full how slow is the camera (imagine a lion taking down a zebra - may last more than 3-4 seconds, I would hate to be limited to only 1 to 2 FPS after the buffer fills).
RichFisher wrote:
There are really 2 questions here. Yes hope much can you shoot at max speed but also once the buffer is full how slow is the camera (imagine a lion taking down a zebra - may last more than 3-4 seconds, I would hate to be limited to only 1 to 2 FPS after the buffer fills).
And that's what my test shows - with the Lexar 1000x, 4fps.
Lars Johnsson wrote:
I belive you have the perfect cards for your bodies and shooting Jim You will not gain much in buying the fast 1000x cards
Thanks, Lars! I try to keep up on what is current, as well as what applies to the older gear -- it's a lot to remember. So far, you have been right on memory cards and a lot of other gear. We all appreciate your sharing your test results and experience.
RichFisher wrote:
how do you tell the difference between the 2 versions. All my cards say 90 MB/sec and UDMA 6. B&H shows a card which still claims 90 MB/sec but claims UDMA 7. Still 90 MB/sec. In the real world what is the difference?
Different label - wonder if they are any faster once inside a 1Dx? I can get a burst of ~50 shots with either the Lexar 1000x or Extreme Pro (UDMA 6) card before it noticeably slows done.
RichFisher wrote:
I then calculated how fast the camera can shoot when the buffer is full.
Now we're talking. That is useful information for those who feel they need the ability to shoot very long bursts. How fast will the rate be before filling the buffer (in most cases, the specs tell us roughly) and once the buffer fills what will the rate drop to.
I think this post is awesome, and it showcases some great numbers for people who need to know what card/how many shots they can get in relation to the speed of the card.
Thanks Lars!
gdanmitchell wrote:
Now we're talking. That is useful information for those who feel they need the ability to shoot very long bursts. How fast will the rate be before filling the buffer (in most cases, the specs tell us roughly) and once the buffer fills what will the rate drop to.
Dan
Glad to help. I guess the high school algebra is useful after all
The after buffer is full is the information I value most. Before the buffer is full, I will get 12 FPS with any card, but after that cards matter.
Gunzorro wrote:
Thanks Lars! It's nice to have some practical yardstick to gauge these newer cards. It would be nice if we had a summary for which CF/SD cards are most appropriate to which camera bodies, from say 1Ds2 on up through 1DX/5D3/T5i models. I gave you a like on your last post, but it applies to the whole thread.
Dan -- I'm surprised you don't "get" the reason for the testing procedure (30 sec) to reduce time variables that would crop up in tabulating exposures in shorter bursts. While the 30 sec does have some application for action shooters, it is being used to quantify test results, not as propaganda encouraging "spray and pray" technique. Once again, you've turned this into a treatise against gear acquisition and individual testing.
You will not gain much in buying the fast 1000x cards