will the sandisk extreme pro 95mb/s still hit the 6D's camera buffer in RAW?

as above - i just purchased the mid-tier version (45mb/s) for use on my 6D, in RAW-only mode. disappointed to see that it is still not fast enough to prevent hitting the buffer.

any idea if the 95mb/s will perform well and maintain the fps rate?

thanks!

I don't think it is possible in any camera to not hit the buffer limit.

the.forumer wrote in post #16992919
any idea if the 95mb/s will perform well and maintain the fps rate?

I'm not sure about the 6D, but I know that with the 50D and 7D card speed made no difference to the number of shots to fill the buffer. On the 7D I tried a 60MB/s card and a 1.5MB/s card - both filled the buffer with the same number of shots.

mine1 wrote in post #16993026
I don't think it is possible in any camera to not hit the buffer limit.

I'm pretty sure this is true if shooting raw. With jpegs it may be a different story.

Sure, the 6D is slower, but the 1DX at full speed produces over 350 MBytes/s. Obviously the cards you are looking at are far from that.

Regardless of this, it has been shown that several camera models don't write to the card whilst shooting when using RAW. So a faster card allows quicker recovery, but not longer bursts. As stated above, that probably is true still for all of them, although I don't know for sure.

apersson850 wrote in post #16993441
As stated above, that probably is true still for all of them, although I don't know for sure.

My ancient memory may be playing up again, but I seem to recall that somebody tested the 5D MkIII and came to the conclusion that it did write to the card while writing to the buffer and that card speed did affect buffer capacity. But I may be mistaken.

Starting with, IIRC the 5Dc and maybe the 30D, Canon was touting "write-through" buffers, where the buffer would be reading the sensor while writing to the card.

Jon wrote in post #16993971
Starting with, IIRC the 5Dc and maybe the 30D, Canon was touting "write-through" buffers, where the buffer would be reading the sensor while writing to the card.

Well, it definitely didn't implement it on the 50D or 7D when shooting raw.

I don't have a 5D3, but on my 1DX switching between a Transcend 400x card and a Lexar 1000x card yields approximately 20 more RAW files before the camera slows. On other cameras that I have tried with fast vs (relatively slower cards) the difference is less with cameras such as the 40D and 1D3 only showing 2-4 frames difference, though with the faster cards the buffer does clear much faster.

hollis_f wrote in post #16994361
Well, it definitely didn't implement it on the 50D or 7D when shooting raw.

With my 5DII card speed can make a big difference. With the latest generation of Sandisk Extreme (gold label) the camera never stops shooting, it just slows down. With my older Ultra/Ultra II/Extreme III cards it will outright stop, then take a shot, then stop, repeat.

I recently bought a Sandisk Pro 95mb/s SD card to see if it would improve upon the performance of the 45mb/s Sandisk version in my 6D. Made no difference whatsover, I tried full raw, small raw, raw & jpeg etc. No differnce whatsoever. I tried turning off any of the in-camera features that improve images etc - still made no difference. I will stick to 45mb/s if I need more cards in the future.

I do not know how the buffering works with Canon but I do many years of experience in IT and know how storage works. Camera takes a continuous sequence of pictures, uses buffer and starts saving the pictures to a memory card. The camera now has data to move the card and the speed of the card determines how fast the data will move. Slower cards would create slower write speeds and cause a bottleneck of the data transfer.

Try using 2 very different cards for a controlled test. 15mb/s versus 45mb/s would be easier to tell a difference when using 2 super fast cards. You can also try shooting in JPG and see if that makes any difference. The camera processes RAW file to be displayed as JPG on the LCD and this is done in camera. Could it be that the buffer of the camera is used for capture of the RAW file as well as extra processing to add JPG data to the same file.

I've done an experiment comparing the number of raw shots it takes to fill the buffer of the 7D when using various speed CF cards. Here's the thread describing the experiment - LINK

For those who don't want to read the whole thing here's the results.

The way it was explained to me is that the camera(s) can only hold so much data in the buffer regardless of the speed of the card. The download speed from the camera to the computer, however, will be faster with the faster cards (I'm told). Pretty much the difference is really negligible so it seems the 45MBs are good enough. Really not going to make any difference in the buffering in-camera.

There is a huge difference on the 5D III if you use a slow card or a fast card.

See the link below

http://robgalbraith.co​m/index-2.html

Select "CF/SD/XQD" and then choose a camera model

Unfortunately, Rob doesn't measure the number of shots before the buffer fills. He measures the number of shots in 30s. So the camera (1) fills the buffer, then (2) starts writing to the card until there's space in the buffer, then (3) takes more shots - goto (1). Obviously, the faster it does step (2) the faster it gets to step (3) and the faster it can repeat.

But it gives you no idea at all as to how many shots will fill the buffer.