Thoroughly confused - cards for Z8

If you would like to post, you'll need to register. Note that if you have a BCG store account, you'll need a new, separate account here (we keep the two sites separate for security purposes).

RickW

Well-known member
Supporting Member
Marketplace
I’m a photographer, not a tech geek and bought a Nikon Z8 with an eye towards transitioning away from my beloved dslr’s. I was under the impression that I needed to invest in CE cards and went out and spent a lot of coin on a bunch of 256 CE cards. I shoot a lot of sports, therefore a lot of images per outing. Now I think I just read that I could have used my gigantic collection of big capacity XQD cards accumulated while shooting with my Nikon DSLRs. Does using XQD cards in Z8 diminish the shooting/editing experience? Again, I’m not all that interested in theoretical upload/download speeds or buffers (cannot ever recall exceeding the buffer capacity on any shoot). A couple of moments “longer” using XQD is a price I am willing to pay if it permits me to stop wasting money on additional CEcards. Are XQD cards useable in the z8 and 9?
 
yes they are usable but will limit any burst shooting to short bursts and it will take longer to clear the buffer. but if it’s not a problem for you the. it’s not a problem

that said most of us shooting more action oriented subject matter tend to benefit from cfe-b cards as they have the potential of being dramatically faster. this results in the ability to shoot longer bursts, maintain a reasonably high frame rate even when you fill the buffer and the buffer clears dramatically faster

i like the delkin cards. the black model is their faster card, their power model is their more general purpose card, however if you get power cards at 640gb and above the performance goes up dramatically

older cards like the lexar and sandisk are known to run hot and i would avoid them
 
I'm using some of my old, but not well used, XQD cards on my Z8. No problem at all. But if you're going to do 20fps for more than a couple of seconds or repeated in a row you'll run out of buffer room. But for the most part not a problem. Last weekend I had enough photos that I needed to use my SD card in the second slot, UHS II. Works just fine and can even do 20 fps but you need to let it flush the buffer more often.
 
yes they are usable but will limit any burst shooting to short bursts and it will take longer to clear the buffer. but if it’s not a problem for you the. it’s not a problem

that said most of us shooting more action oriented subject matter tend to benefit from cfe-b cards as they have the potential of being dramatically faster. this results in the ability to shoot longer bursts, maintain a reasonably high frame rate even when you fill the buffer and the buffer clears dramatically faster

i like the delkin cards. the black model is their faster card, their power model is their more general purpose card, however if you get power cards at 640gb and above the performance goes up dramatically

older cards like the lexar and sandisk are known to run hot and i would avoid them
I appreciate the answer, but what I’m trying to learn is what “dramatically “ means. Now that I’ve learned that xqd cards work, I imagine my best course of action is to just go out and experiment with the xqd’s and learn for myself.
 
I appreciate the answer, but what I’m trying to learn is what “dramatically “ means. Now that I’ve learned that xqd cards work, I imagine my best course of action is to just go out and experiment with the xqd’s and learn for myself.
dramatically means xqd cards maxed out at about 400MB/s. general purpose (non performance oriented) cfe-b cards are like 800-1000MB/s and performance oriented cfe-b cards tend to be between 1400-1600MB/s.

you can work out what this means in practice by using file sizes. a z8/z9 lossless compressed raw is about 65MB.

so 400MB/s* / 65MB per image = 6fps

800MB/s / 65MB = 12fps

1300MB/s / 65 = 20fps**

likewise if we think about buffer clearing, if the buffer can hold 30 frames we can estimate how long it takes to clear the buffer:

30 frames / 6fps = 5 seconds

30 frames / 12fps = 2.5 seconds

30 frames / 15fps = 2 seconds

30 frames / 20fps = 1.5seconds**

so real world means if you shoot a burst with an xqd card, you may be waiting for as much as 5 seconds before you can shoot another burst. with a fast card you can expect half that or better.

* not all xqd cards can actually achieve 400MB/s
** as we get near 20fps other parts of the camera start factoring in, and we cannot sustain that frame rate in the real world at this time. i think the most the camera can sustain at this time with lossless compressed raw is about 15fps.
 
Last edited:
I appreciate the answer, but what I’m trying to learn is what “dramatically “ means. Now that I’ve learned that xqd cards work, I imagine my best course of action is to just go out and experiment with the xqd’s and learn for myself.
The opposite question is what do you mean by "Are XQD cards useable"? Which depends on how you shoot. When I got my Z9, I did buy a CFExpress card and put it in slot 1 and one of my XQD cards in slot 2. Didn't really notice a difference (no overheating, could shoot for 20fps raw and not notice a slowdown in the couple of seconds I typically shoot, did video). Eventually, as prices have come down on the CFExpress cards, have bought more (the first was the most expensive). Figured if I am invested in a new camera and new glass, why not also get the best possible card for these cameras. But if you have lots of XQD cards, and prefer not spending the money, for sure they are "useable".
 
dramatically means xqd cards maxed out at about 400MB/s. general purpose (non performance oriented) cfe-b cards are like 800-1000MB/s and performance oriented cfe-b cards tend to be between 1400-1600MB/s.

you can work out what this means in practice by using file sizes. a z8/z9 lossless compressed raw is about 65MB.

so 400MB/s* / 65MB per image = 6fps

800MB/s / 65MB = 12fps

1300MB/s / 65 = 20fps**

* not all xqd cards can actually achieve 400MB/s
** as we get near 20fps other parts of the camera start factoring in, and we cannot sustain that frame rate in the real world at this time. i think the most the camera can sustain at this time with lossless compressed raw is about 15fps.

likewise if we think about buffer clearing, if the buffer can hold 30 frames we can estimate how long it takes to clear the buffer:

30 frames / 6fps = 5 seconds

30 frames / 12fps = 2.5 seconds

30 frames / 15fps = 2 seconds
Very much appreciate the details and your time laying it out. But, as I said in the opening of my OP, I’m no tech guy and really not sure I understand what it means. Fault is mine, not yours. I am just gonna go find some action, load my xqd card and try it out to see for myself in terms I can easily digest. Buffer rates, transfer speeds and other tech specs etc are meaningless to me. If I go and shoot a session and find I can’t achieve my desired 20 fps or a “full buffer” prevents me from shooting the burst I want, I will quickly understand the physics.

Again - many thanks to those trying to help, but I fear that only hands on experience is going to help me.

Added - I won’t actually “understand” the physics, but I will understand what I might or might not be giving away by using the old cards
 
Using xqd cards on a Z8/9 is like installing a speed/rpm limiter in a sports car. Yeah you can do it and the car will drive but it will not be able to do what it was designed to do which is drive fast

The card speed has to do with how long it takes to get the image off the buffer and into the card. Slower cards can't keep up with the camera frame rate. The camera stops shooting when the buffer is full.
Here is an article by Thom Hogan discussing card options:

 
...The camera stops shooting when the buffer is full...
Well... technically yes. But in practice it slows down. What sort of fps you get once the buffer is stuffed depends on the card speed.

I did a good bit of testing with the Z9 when I first got it. With XQD cards at 20fps I was getting about 30-32 frames before the frame rate slowed down. I never measured exactly what fps it dropped to but somewhere in the 4-6fps range. With a Delkin Power CFE I got about 80-85 frames before frame rate slowed to about 15fps. There is a huge difference in how fast the buffer clears. A couple of seconds for the CFE and 10-15s for the XQD.
 
Well... technically yes. But in practice it slows down. What sort of fps you get once the buffer is stuffed depends on the card speed.

I did a good bit of testing with the Z9 when I first got it. With XQD cards at 20fps I was getting about 30-32 frames before the frame rate slowed down. I never measured exactly what fps it dropped to but somewhere in the 4-6fps range. With a Delkin Power CFE I got about 80-85 frames before frame rate slowed to about 15fps. There is a huge difference in how fast the buffer clears. A couple of seconds for the CFE and 10-15s for the XQD.
A real world difference like that IS something I can wrap my head around and it would indeed matter. Going to see for myself.
 
Last time I asked the question most folks here were recommending saving to only the fastest card and using the other slot as an overflow. I was more in the belt and suspenders mode wanting to write an instant backup to card 2, but more folks here were like, don't bother, the chances of failure are very slim.
 
Last time I asked the question most folks here were recommending saving to only the fastest card and using the other slot as an overflow. I was more in the belt and suspenders mode wanting to write an instant backup to card 2, but more folks here were like, don't bother, the chances of failure are very slim.
yah, i save only to the fastest card and then manually copy images to the sd card if i feel it is necessary in order not to be impacted by the slow speed of the sd card.

in reality this is also an argument for the fastest media you can get even with the z9, because it _appears_ (at least right now) that writing to the second card "counts against" your fps*.

* so for example, if your cards can sustain 15fps, if you only write to one card, you can sustain 15fps. but if you write to both cards, you'll probably only be able to sustain 7.5fps.
 
Design a little test of your own to compare card performance using settings you routinely use.

I did a test of seeing how many images I could capture using common shooting settings to my cards before the buffer showed 0 and the FPS slowed. When I compared the performance of a Prograde Cobalt 165GB CFE, a SanDisk Extreme 128GB CFE, and a Sony 128GB XQD here is what I found

Shooting at 20fps
ProGrade Cobalt CFE- RAW Lossless Compressed 75 to 0 buffer or about 3.75 seconds, RAW HE* 160 to 0 buffer or about 8 seconds
Sandisk CFE - RAW Lossless Compressed 50 to 0 buffer or about 2.5 seconds, RAW HE* 90 to 0 buffer or about 4.5 seconds
Sony XQD - RAW Lossless Compressed 30 to 0 buffer or about 1.5 seconds, RAW HE* 50 to 0 buffer or a about 2.5 seconds

These number are not exact but the difference is significant enough to show differences in card performance.

These tests only take into account shooting stills and don't address video and video heat performance.
 
Back
Top