I took the plunge... Nikon Z8 and now I need help.

[David H. Hartman]

good brands include Delkin (Black, Power) and Prograde Digital (Cobalt in my case).

Thank you for this advice. I don't know if I can scratch together enough cash to buy a CFExpress at this time. I just read yesterday memory prices are going up.

I just bought a RRS L-Plate for the Z8. I should have waited.

I find SD more practical in the field but I would like to save images to both SD and CFExpress at the same time. That way if my SD card tanked I'd still have my images. Is that a good strategy?

Thank You,

Dave

[David H. Hartman]

I'm looking at a ProGrade Digital 240GB CFexpress 4.0 Type B Gold with speeds of Max 3100/1600 MB/s Read/Write (Min Sustained Write Speed: 700 MB/s). I think I can swing the price of this one. I've got to do some counting. At today's CC interest I can't afford to over spend.

Dave

[Les Olson]

What use do you need the high speeds for?  High speeds give you more buffer at high frame rates and are necessary for better video, but if you don't regularly fill the buffer and don't do video you may not need a fast card.

SD cards have a life expectancy. Quality and storage conditions affect life expectancy but you should get five years or 3,000 read/write cycles, whichever comes first. Failure is sudden, so if you keep using cards until they die, you are likely to be inconvenienced. A better strategy is to buy new SD cards every time you buy a new camera or every five years, so that the risk of failure is low.

[David H. Hartman]

I hope faster cards are less stressed, run cooler: maybe/maybe not? Transfers to a computer takes less time.

I dabble in video. I want clean, sharp video at 60hz. I have spent zero time exploring the Z8's video capabilities at this time.

Better cards have better warranties. The longer the warranty period the more confident the maker is of the quality of their product (or so I surmise).

I've never had an SD card fail but I've had a Crucial 0.5TB 2.5" Internal SSD fail. It gave slight warning. I managed to back it up to an HD then I tried macOS disk utility> firstaid> in recovery mode and then Diskpart in Win10 CMD and nothing helped. The backup to an HD was the last time I could read from the drive. I was surprised that the backup completed.

---

Thanks for the advice on SD cards.

Dave

[David H. Hartman]

He who hesitates saves money.

I placed an order for a ProGrade Digital 240GB CFexpress 4.0 Type B Gold Memory Card ($25.00 off) and a ProGrade Digital CFexpress Type B & UHS-II SDXC Dual-Slot USB 3.2 Gen 2 Card Reader ($12.00 off). I think the sale started Saturday.

Oops! I did some cut and paste and accidentally deleted that the supplier was B&H Photo-Video, NY; my favorite supplier since Lee-Mac Camera in Pasadena closed I guess a couple of decades back.

Dave

[David H. Hartman]

My first Nikon Z8 would not give up the CFexpress card. I inserted the card once and that was it. No amount of finesse would free the card. I returned the camera to the seller and yesterday got the replacement. All seems well with this one.

This seller only offered SanDisk CFexpress B cards, no other brands. I bought a 128GB SanDisk card that identifies as a "ProGrade CFexpress PG05.5 Media" in macOS Disk Utility?? I wonder if SanDisk or whatever the card really is will give over-heating problems? Perhaps the over-heating problems have been corrected.

Now I'm scouring the internet for articles and videos of various photographer explaining how they setup their cameras with particular interest in how they use the advanced auto  focus features.

Dave

[pluton]

The general consensus is that The Sandisk CFe cards are more likely to overheat than the higher data rate cards such as Delkin Black, Nextorage Pro, and Prograde.  Heating (and overheating) is said to occur when a large amount of data is being pushed to the card, such as 4K and 8K video.  A look at the data rates of the various frame rates and codecs will give you an idea of when overheating (card or camera) might be more likely.  The camera electronics heat up also according to the amount of data processing that is being undertaken.  Also, the camera batteries heat up a bit when being heavily drained.
Some extreme stills shooters (20FPS for hundreds/thousands of shots in a short time period) report hot card warnings.  Camera heating is reported separately from card heating in the Nikon Z8 system.
I got my Sandisk Extreme Pro CFe Type B 128GB card to overheat at about 12 minutes of recording 4K ProRes 422 10-bit, 30FPS,  at room temperature.  I haven't made the same test with the ProGrade cards yet.

[Les Olson]

No such thing as a free lunch, as they say: if you want to shift data you have to provide power. It is impossible to convert all the power you provide into useful work, and the left-over power is dissipated as heat. Power dissipation in watts = (current in amperes)squared X resistance in ohms, so higher current has a big effect on power dissipation.

Current for a CF Type B card varies from 1500 mA to 3000mA, so at 3.3V power consumption = amps x volts = 5-10 W. That is a lot of power, and a lot of heat as a result, and what you are looking for is a card with low power consumption relative to its size and speed. Power consumption is substantially higher for higher capacity cards, so a 256 GB card is more likely to overheat than a 128 GB card, and is higher for reading than writing, so make sure the data is apples to apples.

The card manufacturers have not publicised power consumption data, which leads me to surmise there is not much difference. 

[Ilkka Nissilä]

No such thing as a free lunch, as they say: if you want to shift data you have to provide power. It is impossible to convert all the power you provide into useful work, and the left-over power is dissipated as heat. Power dissipation in watts = (current in amperes)squared X resistance in ohms, so higher current has a big effect on power dissipation.

Current for a CF Type B card varies from 1500 mA to 3000mA, so at 3.3V power consumption = amps x volts = 5-10 W. That is a lot of power, and a lot of heat as a result, and what you are looking for is a card with low power consumption relative to its size and speed. Power consumption is substantially higher for higher capacity cards, so a 256 GB card is more likely to overheat than a 128 GB card, and is higher for reading than writing, so make sure the data is apples to apples.

The card manufacturers have not publicised power consumption data, which leads me to surmise there is not much difference.

This is not at all correct; CFexpress type B card temperature behavior varies greatly between cards and it has significant ramifications on the cameras' usability. When it came clear that overheating was a problem (mainly with the Canon R5), tests have been made of the temperature the cards run in when the same task is carried out with different cards (using a specific camera). The published temperatures also reflect the performance under extended stress, i.e. cards that run cooler tend to maintain their speed for a longer time in burst shooting or high data rate video recording. Here is one such site:

https://www.thessdreview.com/our-reviews/cfexpress/delkin-black-cf-express-type-b-512gb-memory-card-review/4/

Unfortunately these tests do not include the newest cards. For example, Prograde Digital Cobalt (which is their high-performance card) has been superseded by Iridium. Nexstorage is another high-performance card manufacturer (started up by former Sony engineers, reportedly) that hasn't been included in those tests. I would stick to Prograde Digital or Delkin cards (both are from time to time discounted heavily at B&H) when purchasing CFexpress type B cards.  There are other tests which did not record the temperature of the card but simply the burst performance or video recording times before overheating under various conditions. For the Z8, for example, Gerald Undone and Ricci Chera have published some overheating tests. It turns out the card type greatly affects how long the camera can record high-quality video before heating becomes a problem.

[David H. Hartman]

Thank you for all the responses.

I wish the seller of the camera offered Delkin Black but they didn't. I had to buy a CFexpress card from the seller of the camera so if bad luck struck twice the return would be easy. If the SanDisk/ProGrade 128GB card run warm but not excessively I'll probably slide it in my D850. I'm thinking the data transfer speeds of the D850 are slower so heat might be less of a problem with it.

Thanks again, you guys are great!

Dave

[pluton]

Dave, After reading the last few posts, I ran a test with my Z8 + Sandisk 128GB CFe-B card.  I had run a test a few months ago recording using the data-heavy ProRes 4:2:2 10-bit format at UHD3840/30FPS.  It ran about 12-14 minutes before issuing hot card/hot camera warnings.  Just now, I tested again, same camera and card, but recording H.265/10-bit at regular HD1080/30FPS and it went so long...over 80 minutes...that I shut it off out of boredom.  The card and camera got slightly warm, not hot, and no warnings from the Z8.  Conclusion:  Low data rate recording formats (like regular HD 1080) generate less heat than the high data rate formats such as ProRes.

[Les Olson]

This is not at all correct; CFexpress type B card temperature behavior varies greatly between cards and it has significant ramifications on the cameras' usability. [...] It turns out the card type greatly affects how long the camera can record high-quality video before heating becomes a problem.

If the card gets hotter than ambient temperature, the only place that extra extra heat can come from is power leaving the battery and being dissipated as heat. One card cannot get hotter than another unless it consumes more power.

The hottest card on the site you linked to was just over twice as hot as the coolest, and as I pointed out, manufacturers' data shows that for the same brand changing the task and the card size can double the power consumption, and double the power consumption means double the final temperature. A lot of the reported differences in final temperature are much smaller - eg, 40 degrees above ambient vs 30 - and that corresponds to a smaller difference in power consumption than you would expect from changing card size.

That means that testing how hot cards get requires being very careful, and a lot of people are not nearly careful enough (eg, not using the same size cards). However, allowing that some cards get hotter than others, the question is whether the reason some cards get hotter is, as everyone seems to assume, differences between the cards, or differences between how the cameras handle the cards. Eg, it has been suggested that some cameras allow and others don't allow cards to go to standby or sleep immediately after a burst, which may matter because power consumption is very low in standby and sleep and the card has a chance to cool down.