new CF express cards from Delkin on Z9 buffer test

[jmurthy]

If anyone is curious, got the new delkin CF express card 512 GB at USD 499.99.

I shot a burst of lossless compressed RAW (not HIE ) to the card at 20 fps.. after about 350 frames , the buffer slowed slightly for a second or two and i let go of the shutter release. It started right back thereafter when i pressed and held the shutter release aagain. I was shooting in CAF animal detect at SS 1/2500 ISO 500 f/5.6

 

[jcgamble]

Was that the Z9?

 

[Ivan Rothman]

With the new 150GB Delkin card I couldn't get to more than about 85 images before the camera started to stutter.

 

[jmurthy]

update: tried the test again today. consistently getting 240-300 frames on Z9 lossless compressed raw ( no jpeg) at 20 fps CAF with tracking, animal eye detect, f/5.6, ISO 2000, ss 1/2500.
buffer takes 1-1.5 seconds to completely clear out for the next burst.
Its fantastic! will prob never use these many number of frames at 20fps. good to know that its completely “over engineered” for my style of shooting.

 

[EricBowles]

That's great - thanks for sharing your results. The 240-300 frame bursts are fantastic, but what is really jaw dropping is clearing the buffer with a short hesitation and being able to launch a new burst of another 240-300 frames.

 

[Ivan Rothman]

It's interesting that you are getting such good results with your Delkin Black 512GB card. I and some others have reported that we are getting less buffer with the new but smaller Delkin Black cards, in my case with the 150GB. So it well may be that the larger card is also a faster card. Delkin had originally reported much better results with the new card in the Z9 than any of us were seeing - there is a thread here about that. Maybe Delkin did its testing with the larger size card. I wish I had the larger size card so that I could do comparison testing of the 2 cards.

 

[jmurthy]

You can only see EXIF info for this image if you are logged in.

The card on the left gives me 79-80 frames with lossless raw compressed before the buffer runs out. the one on the right is the newer faster card that i am very happy with.

 

[arbitrage]

Have you checked that it is really getting 20FPS the entire time? Every other test I've read online for the new Delkin Black card (even the 512GB one) are only getting ~85 shots. Then the camera slows to about 15FPS and continues to shoot fairly steadily. Delkin even rewrote their marketing after many complained. If you are getting a sustained, consistent 20FPS for 300 shots then you have a card outperforming the others.

 

[Alistair]

That's great - thanks for sharing your results. The 240-300 frame bursts are fantastic, but what is really jaw dropping is clearing the buffer with a short hesitation and being able to launch a new burst of another 240-300 frames.

I agree. I find this to be the much more useful aspect of buffer performance - how many and how soon the next burst can be.

 

[jmurthy]

Have you checked that it is really getting 20FPS the entire time? Every other test I've read online for the new Delkin Black card (even the 512GB one) are only getting ~85 shots. Then the camera slows to about 15FPS and continues to shoot fairly steadily. Delkin even rewrote their marketing after many complained. If you are getting a sustained, consistent 20FPS for 300 shots then you have a card outperforming the others.

that seems odd. i will check again tomorrow on all my settings to confirm. traveling today. I hope i am not making a rookie mistake of not looking at some setting deep in the menu. but all cam settings are right and shooting at 20fps.

 

[Dave Stargazer]

@jmurthy glad you are happy with the performance It is pretty amazing really.

For reference, here is my previous post about my tests of that card (Delkin Black 512gb):
https://bcgforums.com/index.php?threads/new-delkin-black-claims-no-buffer-on-z9.15455/post-175687

And Delkin's post of my findings:
https://www.delkindevices.com/product-help/

If you want to dig deeper, look at the millisecond time stamps on the frames in the exif (which is what I did). Also have a 'live' subject so the camera has to react, and spend time processing/following that (it def has an effect) - lens cap or stationary camera will get better, but not 'real world' results. I should post some of my graphs haha - wasn't sure if Delkin would post those or not. Similar to arbitrage, Ivan, etc. for lossless, I got 80ish frames at 20fps followed 15fps (variable), which is still great. HE* was nearly 20fps continuous, and HE was even closer to 20fps (but not quite 100%).

Bottom line for me is I just keep shooting with the knowledge that the card keep up

Cheers!

...Dave...

 

[EricBowles]

that seems odd. i will check again tomorrow on all my settings to confirm. traveling today. I hope i am not making a rookie mistake of not looking at some setting deep in the menu. but all cam settings are right and shooting at 20fps.

It can be a bit tedious, but my testing methodology involved scrolling through the time shot and recording the number of images in each full second interval (just using the EXIF data in playback mode on the camera). This approach let's you see that your initial burst is at the full 20 fps, and the relative impact once the buffer starts to fill. I also monitored the Remaining Images countdown to get an idea of how many shots where in the buffer - from the peak to whenever you stop your burst.

 

[Dave Stargazer]

Ok, here's one of my graphs - Lossless F 70-200mm f/2.8 on the Z9 2.0 firmware

Perfect performance would be a flat line at the bottom - the higher the dot, the longer the lag for that frame. This is excellent performance imo.

Btw, Eric's methodology works to do a decent assessment, and doesn't involve semi-programming to extract and clean the exif info, etc. haha

Cheers!

...Dave...