Again, thanks to all for the comments on my earlier question. I reply here to
Larry’s comment as his comment provided me with a template for what follows.
Any credit is his, any blame is mine…
Three SDXC cards. One K-1 with two slots.
Card A = Lexar Professional SDXC II U-3 150MB/s - 1000x
Card B = Lexar Professional SDXC I U-1 95MB/s - 633x
Card C = SanDisk Extreme SDHC I U-3 90MB/s
Questions:
- does it matter which card I use in the K-1?
- does it matter if I mix and match e.g. A in Slot 1 and C in Slot 3?
Condition: K-1 in Manual mode, Hi-speed Continuous mode. Shooting in Liveview.
Macro lens focused on face of Casio digital watch.
Settings: RAW (DNG) files written to Slot 1,highest quality jpeg written to
Slot 2
Variations: Slot 1 only vs. Slot 1 + Slot 2
Parameters:
- Time to fill the buffer
- # of images to fill the buffer
- Time to clear the buffer once it is full
Results: It doesn’t matter. Mostly.
With one card only, in Slot 1, using Cards A, B, or C, the K-1 took roughly* 3
seconds to fill the buffer. That was after 13 images. Once the buffer was full,
in about 12 seconds the LV screen reappeared and it was possible to take one
shot every second or two. Assuming no additional shots beyond the original 13,
an additional 23 seconds required to clear the buffer.
With second card in Slot 2, results are different. With A + A, A + B, A + C, B
+ B, B + C, or C + C, the K-1 took approximately 2 seconds to fill the buffer,
11 images filled the buffer, and clearing the buffer took about 35 seconds
after that initial burst of 11 shots.
So, no appreciable difference among these (relatively modern) SD cards, despite
their somewhat different self-proclaimed ratings.
But quite a difference when shooting with cards in both slots compared to a
single card in Slot 1 written to with DNG files!
*Caveat: The watch I targeted with my macro lens (to get the timings) was in
normal mode, showing time to the nearest second. I may re-do some of this
tomorrow with the watch in Stopwatch mode showing hundredths of seconds. I am
sure there was at least some slight variation among the cards, but it was just
too small to be picked up with the approach I used.
You may now return to your regularly programmed event…
stan
> On May 29, 2018, at 1:52 PM, Larry Colen <[email protected]> wrote:
>
>
>
> Stanley Halpin wrote on 5/29/18 10:13 AM:
>
>> Question: Is there any reason not to mix-and-match? I pretty much always
>> have cards with the same speed rating in both card slots, but I don’t know
>> if I am being a bit too compulsive…
>> And I would guess that if I did mix them, that it would make sense (?)
>> for the faster card to be placed in the slot where the larger RAW files are
>> being written.
>
> It is my understanding that at those rates it is not the card speed that is
> the limiting factor. If you're driving down a heavily patrolled freeway and
> you have your choice of a Corvette, a Ferrari or a McLaren it doesn't really
> matter if one tops out at a 178 MPH, another at 167 and the third at 153,
> with all those cops, it's hard to do much more than 100 for very long.
>
> In a similar sense, the limiting factor in the camera with those cards is the
> bus speed. I don't remember the details, but I don't think the camera can
> maintain much better than about 60Mbps. It's further complicated by some
> cards (because of buffering?) can store a little bit of data at a much higher
> rate than they can store data at a sustained rate, and it's that burst rate
> that is usually advertised. There are other ratings (for video) that list
> the sustained rate that they can store data.
>
> In short, if you only shoot a few photos at a time, then the burst rate is
> more critical, if you shoot action and take photos fairly continuously, then
> the sustained rate is more critical.
>
> In your case, the way to test would be to put the camera in high speed motor
> drive mode, hold the shutter down, count how many frames you can take until
> the buffer fills, then measure how long until the "write" light goes out. If
> there is no substantial difference between each of the cards, it doesn't
> matter.
>
>> Thanks!
>> stan
--
PDML Pentax-Discuss Mail List
[email protected]
http://pdml.net/mailman/listinfo/pdml_pdml.net
to UNSUBSCRIBE from the PDML, please visit the link directly above and follow
the directions.
