Nikon D750 & D780

Dumb question.

I'm borrowing another camera for a shoot tomorrow d7100 - i assume this is applicable to most of the upper end of the nikon range.

I have a 64gb card in slot one
I have a 32gb card in slot two

Camera is set to Raw slot 1, JPG slot 2

If i just have the 64gb card in it says 1.1k shots remaining. If i add the 32gb card to slot 2 it says ~520 shots remaining (i assume it always gives the lesser amount)

Question 1)
Are raw and jpg similiar file sizes

Question 2)
What happens if i run the shots remaining counter to 0. 32gb card will be full, but space left on the 64gb card - i assume it will just write to card slot 1 and let me shoot still?
That seems odd, I would expect far more jpegs than that from a 32gb card. According to google the D7100 jpeg file sizes are 8.2mb therefore you'd expect to fit circa 3900 jpegs onto a 32gb card. I would double check your settings, and also make sure that the 32gb card is empty.

As for question 2, if set to slot 1 RAW and slot 2 jpeg backup on my D750 it would stop shooting when either card reached 0 and wouldn't automatically write both RAW and jpeg to the available space on the other card.
 
Question 1 = No, RAW is larger

Question 2 = not certain, never reached that state. It has so many variables/imponderables. For a start, will it just carry on shooting RAW? I should imagine the 32 GB card will remain with spare capacity (the RAW will be more than double the JPG size)

Others may have more experience; mine is just based upon hypothesis (RAW is larger than JPG though for certain)

That seems odd, I would expect far more jpegs than that from a 32gb card. According to google the D7100 jpeg file sizes are 8.2mb therefore you'd expect to fit circa 3900 jpegs onto a 32gb card. I would double check your settings, and also make sure that the 32gb card is empty.

As for question 2, if set to slot 1 RAW and slot 2 jpeg backup on my D750 it would stop shooting when either card reached 0 and wouldn't automatically write both RAW and jpeg to the available space on the other card.

Card is empty

Here’s a video, hopefully I’m doing something stupid?

View: https://youtu.be/jbyUhYA0VZE
 
Ah I sorted it, it was set to raw only. I assumed setting raw slot 1 and jpg to slot 2 would auto write a jpg but assume as it was set to raw only it only in image quality wasn’t doing jpg.

Changed image quality to raw + jpg and it now says 1.1k shots

Cheers!
 
Camera is set to Raw slot 1, JPG slot 2
Others have answered your question, but I’d add an observation.

Unless you have a specific requirement (for example if you want to had the jpg card to someone immediately after the shoot for their review) RAW slot 1, JPG slot 2isnt really solving any problem. If you need a backup for security / to give you confidence, then make a backup - RAW to both cards. RAW + JPG to different cards is pretty pointless, either you’ll just wipe Card 2 because Card 1 has the RAW files you need, or you’ll have lost the RAW files and only have JPGs.

I’m not saying you need RAW over JPG, but as a backup system RAW + JPG is pointless (IMO of course).
 
havent shared anything for a while, and I haven’t shared any of the paid stuff I do so here’s some shots from a recent wedding :)

Untitled by andyroberts1868, on Flickr

Untitled by andyroberts1868, on Flickr

Untitled by andyroberts1868, on Flickr

Untitled by andyroberts1868, on Flickr

Untitled by andyroberts1868, on Flickr

Untitled by andyroberts1868, on Flickr

Untitled by andyroberts1868, on Flickr

Untitled by andyroberts1868, on Flickr

Excellent images, I know the bride - went to school with her. She seems over the moon with them.
 
Excellent images, I know the bride - went to school with her. She seems over the moon with them.

Oh wow really? What a small world! They were a really great couple to be honest, we did an engagement shoot last year and they took to it really well, so we were always going to get great shots of them on the Wedding day!
 
Ah I sorted it, it was set to raw only. I assumed setting raw slot 1 and jpg to slot 2 would auto write a jpg but assume as it was set to raw only it only in image quality wasn’t doing jpg.

Changed image quality to raw + jpg and it now says 1.1k shots

Cheers!
lol, was just about to post this. It's quite confusing, but as you say not only do you have to set the second card to RAW + jpeg you also have to set it to RAW + jpeg (where there's various jpeg options) in the image quality (y)
Others have answered your question, but I’d add an observation.

Unless you have a specific requirement (for example if you want to had the jpg card to someone immediately after the shoot for their review) RAW slot 1, JPG slot 2isnt really solving any problem. If you need a backup for security / to give you confidence, then make a backup - RAW to both cards. RAW + JPG to different cards is pretty pointless, either you’ll just wipe Card 2 because Card 1 has the RAW files you need, or you’ll have lost the RAW files and only have JPGs.

I’m not saying you need RAW over JPG, but as a backup system RAW + JPG is pointless (IMO of course).
I use raw with jpeg backup and the reason being is for buffer and memory card storage. Let’s be honest it’s very rare a card fails and so for me it’s a case of off setting the risk against reducing speed and buffer, plus having to buy more expensive cards. Each to their own but I’d rather have max performance and at least have some form of back up (which you can still do a lot with) rather than slow the camera down on the off chance I may have a card failure.

Obviously if I was doing paid work I might change this approach, but my gut feeling is that I wouldn’t.
 
Anyone know how to assign min shutter speed to the fn button?

I normally rock auto iso and min shutter 1/200, but I’d like to be able to quickly change the min shutter
I’m not aware that you can do this tbh. I’m assuming you’re shooting in aperture priority and hence the min shutter setting so one quick way would be to shoot in manual mode in darker environments that require shutter less than 1/200, or if you want to stick to aperture priority then set my menu to an fn button and make the min shutter speed option the first option in my menu.
 
I’m not aware that you can do this tbh. I’m assuming you’re shooting in aperture priority and hence the min shutter setting so one quick way would be to shoot in manual mode in darker environments that require shutter less than 1/200, or if you want to stick to aperture priority then set my menu to an fn button and make the min shutter speed option the first option in my menu.
I did try that, but said that it couldn’t be assigned to a quick menu :(
 
Isn’t writing JPEG + RAW slower and results in lower buffer size than writing RAW + RAW?
I've never done a test but I can't see how writing circa 10mb files is slower than writing 30mb ones?
 
I've never done a test but I can't see how writing circa 10mb files is slower than writing 30mb ones?

I don't think it's a question of just writing the files - the RAW is just that, RAW from the sensor but a jpg is derived from that RAW data with all the various settings applied: Colour space, D-Lighting, Picture Control - all 'work' the camera's Processing Engine has to perform before it can write to file. With RAW, this kind of data is just set as a flag for Post Processing - LR, etc.
 
I don't think it's a question of just writing the files - the RAW is just that, RAW from the sensor but a jpg is derived from that RAW data with all the various settings applied: Colour space, D-Lighting, Picture Control - all 'work' the camera's Processing Engine has to perform before it can write to file. With RAW, this kind of data is just set as a flag for Post Processing - LR, etc.
I don’t know tbh, was just my logic. AFAIK it’s not the processing of the camera that affects buffer speed as much as the ability to write the data to the card, and hence why faster cards tend to yield larger buffers. Also RAW files carry a jpeg sidecar anyway so there’s still jpeg processing with RAW files. Would be interesting to know for sure (y) Either way it’s more cost effective for me to shoot jpeg backup, especially now I have the D850 where I’d need the very expensive UHS-II cards as my backup cards :eek:
 
I don’t know tbh, was just my logic. AFAIK it’s not the processing of the camera that affects buffer speed as much as the ability to write the data to the card, and hence why faster cards tend to yield larger buffers. Also RAW files carry a jpeg sidecar anyway so there’s still jpeg processing with RAW files. Would be interesting to know for sure (y) Either way it’s more cost effective for me to shoot jpeg backup, especially now I have the D850 where I’d need the very expensive UHS-II cards as my backup cards :eek:
I can't find similar for the D750 ... but looking at the D500 https://www.cameramemoryspeed.com/nikon-d500/sd-and-xqd-card-speed-test/ shows setting the camera in "Backup" mode (writing to both cards simultaneously) with a Lexar Professional 2933x XQD 64GB (Card 1) and Lexar 2000x UHS-II 64GB (Card 2) - RAW only gives 102 shots (in 30 seconds); JPEG only 261 shots and RAW + JPEG only 72 shots. The D5 was similar but it is down to write speed as when using two slower XQD cards in the D5 the RAW + JPG figures were much closer to the RAW only backup compared to when using two faster XQD cards.

I thought I'd seen similar tests with the D750 but I may be mistaken.
 
I can't find similar for the D750 ... but looking at the D500 https://www.cameramemoryspeed.com/nikon-d500/sd-and-xqd-card-speed-test/ shows setting the camera in "Backup" mode (writing to both cards simultaneously) with a Lexar Professional 2933x XQD 64GB (Card 1) and Lexar 2000x UHS-II 64GB (Card 2) - RAW only gives 102 shots (in 30 seconds); JPEG only 261 shots and RAW + JPEG only 72 shots. The D5 was similar but it is down to write speed as when using two slower XQD cards in the D5 the RAW + JPG figures were much closer to the RAW only backup compared to when using two faster XQD cards.

I thought I'd seen similar tests with the D750 but I may be mistaken.
Either we are saying different things or one of us is misinterpreting the data ;)

Screen Shot 2018-07-18 at 12.33.00.png

So looking at this if you use the second slot as a backup, i.e. you're just using the XQD card you get 166 shots if shooting RAW and jpeg to the XQD, 231 just RAW and 261 just jpeg, which is what you'd expect. The next section shows writing the same files to both cards, so RAW and Jpeg to both cards (ie 4 copies), RAW to both cards and then Jpeg to both cards. Again as you'd expect, shooting both file types to both cards is slower than writing one file type to each card. Then the last section is writing RAW to slot one and jpeg to slot two, so just one file type to each card. Again fairly predictably speed is somewhere in between the top section eg between writing RAW and jpeg to the XQD only and writing RAW and jpeg to the SD card only.

Also, this proves my theory that it is faster shooting RAW to slot one and jpeg to slot 2 (giving 145 shots) than shooting RAW to both slots 1 and 2 (102 shots) (y)
 
Or some of us just lost the plot! (I'm humble enough to admit it)
 
So i had a shoot yesterday and something weird was happening from time to time.

I'd take a picture, then review it and when i zoomed in once the display would go to a load of black and white dots. If i zoomed in again or zoomed out the picture appeared, then when zooming again it would be fine.

The only things that have changed are:
1) Updated the firmware (doubtful)
2) Writing a backup raw to card slot 2

It was very hit and miss when it happened... anyone had anything similiar?
 
Either we are saying different things or one of us is misinterpreting the data ;)

View attachment 130591

So looking at this if you use the second slot as a backup, i.e. you're just using the XQD card you get 166 shots if shooting RAW and jpeg to the XQD, 231 just RAW and 261 just jpeg, which is what you'd expect. The next section shows writing the same files to both cards, so RAW and Jpeg to both cards (ie 4 copies), RAW to both cards and then Jpeg to both cards. Again as you'd expect, shooting both file types to both cards is slower than writing one file type to each card. Then the last section is writing RAW to slot one and jpeg to slot two, so just one file type to each card. Again fairly predictably speed is somewhere in between the top section eg between writing RAW and jpeg to the XQD only and writing RAW and jpeg to the SD card only.

Also, this proves my theory that it is faster shooting RAW to slot one and jpeg to slot 2 (giving 145 shots) than shooting RAW to both slots 1 and 2 (102 shots) (y)

Who made him get his lab coat out again :D:D
 
I don’t know tbh, was just my logic. AFAIK it’s not the processing of the camera that affects buffer speed as much as the ability to write the data to the card, and hence why faster cards tend to yield larger buffers. Also RAW files carry a jpeg sidecar anyway so there’s still jpeg processing with RAW files. Would be interesting to know for sure (y) Either way it’s more cost effective for me to shoot jpeg backup, especially now I have the D850 where I’d need the very expensive UHS-II cards as my backup cards :eek:

Agreed, while technically the camera does need to process the raw data from the sensor to produce the jpeg it has dedicated hardware which is hugely optimised to be able to rapidly produce the jpeg. With the significantly reduced file size it means it can save the file quicker to the memory card compared to the larger raw file.

Reading the discussion above though it did make me wonder if using both cards with raw to one and jpeg to the other would be a quicker way to clear the buffer but reading the table above it seems that's not the case? I was doing a lot of burst shooting at the weekend and the small buffer was irritating, I had the cards set to overflow - does anyone have experience of the buffer speed using two cards compared to one?
 
Agreed, while technically the camera does need to process the raw data from the sensor to produce the jpeg it has dedicated hardware which is hugely optimised to be able to rapidly produce the jpeg. With the significantly reduced file size it means it can save the file quicker to the memory card compared to the larger raw file.

Reading the discussion above though it did make me wonder if using both cards with raw to one and jpeg to the other would be a quicker way to clear the buffer but reading the table above it seems that's not the case? I was doing a lot of burst shooting at the weekend and the small buffer was irritating, I had the cards set to overflow - does anyone have experience of the buffer speed using two cards compared to one?
For the best buffer write to one file type on one card using the fastest compatible card that you can. Writing two files to one card is slower than writing two files split to two cards, ie one file on each card (except if you're writing jpeg only in which case there's no difference).

I rarely ran into the buffer using the D750, but when I did do events that required a large burst the buffer could be annoying. However an easy way to increase the buffer significantly is to shoot lossless 12 bit instead of lossless 14 bit, taking the buffer from circa 15 shots to 25 shots, which is plenty for most occasions. I did some experimenting and could not see a difference in 12 bit and 14 bit files. Admittedly I didn't push the files massively, but certainly will exposure changes with +/- 1.5ev I couldn't see a difference.
 
For the best buffer write to one file type on one card using the fastest compatible card that you can. Writing two files to one card is slower than writing two files split to two cards, ie one file on each card (except if you're writing jpeg only in which case there's no difference).

The post above for the D500 says the opposite though if I'm reading it right, it shows 166 photos for raw and jpeg on a single card but only 145 for raw on one card and jpeg on the other, I assumed that meant it managed a few more shots using a single card.

I rarely ran into the buffer using the D750, but when I did do events that required a large burst the buffer could be annoying. However an easy way to increase the buffer significantly is to shoot lossless 12 bit instead of lossless 14 bit, taking the buffer from circa 15 shots to 25 shots, which is plenty for most occasions. I did some experimenting and could not see a difference in 12 bit and 14 bit files. Admittedly I didn't push the files massively, but certainly will exposure changes with +/- 1.5ev I couldn't see a difference.

I generally only hit buffer issues with motorsports events but I don't do that often enough to justify changing cameras.

John
 
Last edited:
The post above for the D500 says the opposite though if I'm reading it right, it shows 166 photos for raw and jpeg on a single card but only 145 for raw on one card and jpeg on the other, I assumed that meant it managed a few more shots using a single card.
Yes I think I was getting myself confused too now :LOL: Although, it's difficult to say for sure with this test as the XQD is much faster than the SD card, and so it's maybe faster writing both to the XQD than RAW to XQD and jpeg to SD. One thing that confuses me though is that jpeg to SD shouldn't be the limiting factor as writing jpeg only to both cards you get a buffer of 261 so the limiting factor should be the RAW files. Looking at the data as a whole it does look overall that 'splitting' the data to two cards does reduce the speed, so it's clearly not all just down to file size(s). It would be interesting to see the stats for the D750 which uses SD cards for both to see if that has a similar trend.


I generally only hit buffer issues with motorsports events but I don't do that often enough to justify changing cameras.

John
Changing to 12 bit RAW doesn't require a change of camera ;) :p
 
Yes I think I was getting myself confused too now :LOL: Although, it's difficult to say for sure with this test as the XQD is much faster than the SD card, and so it's maybe faster writing both to the XQD than RAW to XQD and jpeg to SD. One thing that confuses me though is that jpeg to SD shouldn't be the limiting factor as writing jpeg only to both cards you get a buffer of 261 so the limiting factor should be the RAW files. Looking at the data as a whole it does look overall that 'splitting' the data to two cards does reduce the speed, so it's clearly not all just down to file size(s). It would be interesting to see the stats for the D750 which uses SD cards for both to see if that has a similar trend.

I would be curious but I suspect the performance difference will be small even if it's in favour of two cards and I find it easier to manage having the files together on one card.


Changing to 12 bit RAW doesn't require a change of camera ;) :p

I only use 12-bit but even with that it's only a 13 shot buffer.
 
I would be curious but I suspect the performance difference will be small even if it's in favour of two cards and I find it easier to manage having the files together on one card.




I only use 12-bit but even with that it's only a 13 shot buffer.
Strange it’s only 13 shots, ‘the web’ quotes 25. What cards are you using?
 
Strange it’s only 13 shots, ‘the web’ quotes 25. What cards are you using?

I was wondering that as well so I did double check I had it set to 12 bit as if it was 14-bit that would agree with the buffer sizes I've seen online but it's definitely 12-bit. It's a Kingston 95MB/s card which I know has a decent write speed as the Sony's accept it for 1000fps and 4K (they're very fussy) although even with a slower Sandisk card I get the 13 shot buffer as well it's just it's slower to drain to the memory card. That's on RAW plus the highest quality jpeg, if I go down a setting to medium or low jpeg I get 15 shots, jpeg only gets 27 shots and raw only gets 16 shots.
 
I was wondering that as well so I did double check I had it set to 12 bit as if it was 14-bit that would agree with the buffer sizes I've seen online but it's definitely 12-bit. It's a Kingston 95MB/s card which I know has a decent write speed as the Sony's accept it for 1000fps and 4K (they're very fussy) although even with a slower Sandisk card I get the 13 shot buffer as well it's just it's slower to drain to the memory card. That's on RAW plus the highest quality jpeg, if I go down a setting to medium or low jpeg I get 15 shots, jpeg only gets 27 shots and raw only gets 16 shots.
Hmmm interesting. I don’t have a D750 anymore to be able to test unfortunately. Shooting 12 bit definitely improves buffer size over 14 bit on the D850 though. Is the Kingston 95mb/s write as well as read?
 
Hmmm interesting. I don’t have a D750 anymore to be able to test unfortunately. Shooting 12 bit definitely improves buffer size over 14 bit on the D850 though. Is the Kingston 95mb/s write as well as read?

I thought it was 95MB/s write but it's actually 80 although I chose it as it was one of the fastest cards in actual tests although it doesn't seem to affect the buffer size as slower cards are the same. I did try changing to 14 bit out of interest and it dropped the buffer from 13 to 11 which isn't as big a change as I expected since online sources suggest switching to 14 bit halves the buffer size.
 
I thought it was 95MB/s write but it's actually 80 although I chose it as it was one of the fastest cards in actual tests although it doesn't seem to affect the buffer size as slower cards are the same. I did try changing to 14 bit out of interest and it dropped the buffer from 13 to 11 which isn't as big a change as I expected since online sources suggest switching to 14 bit halves the buffer size.
80’s not too sluggish. IIRC when shooting 14 bit RAW to slot 1 and fine jpeg to slot 2 I got 12 shot buffer on average using Sandisk Pro Extreme 95mb/s read and write, but as I said I never tried 12 bit. With the D850 using the slower XQD cards changing from 14 bit to 12 bit increases the buffer from around 22 to 35 so would have thought that other cameras show a similar level of increase.
 
This thread needs more pictures as you’re all taking about stuff I don’t understand :p

Well one of the casualties of making excessive use of burst most is a heck of a lot of photos so I'm still trying to summon up the motivation to start digging through them but here's a quick couple:

DSC_8577-L.jpg


DSC_8691-L.jpg


The first one is two of the Nurburgring Ring Taxis, an F-type Jaguar I was in and the Mclaren 720s behind while the second one is me driving the 'Iron Man' Suzuki with its red and gold colouring and IM in the number plate. A friend offered to take photos of the cars for me so I handed over the D750, set it to burst mode, AF-C and F11 thinking that gave a good chance of a photo and he did well given he doesn't use cameras. The second one is a really, really tight crop but still quite usable I thought.
 
I thought it was 95MB/s write but it's actually 80 although I chose it as it was one of the fastest cards in actual tests although it doesn't seem to affect the buffer size as slower cards are the same. I did try changing to 14 bit out of interest and it dropped the buffer from 13 to 11 which isn't as big a change as I expected since online sources suggest switching to 14 bit halves the buffer size.

This website is quite good for comparing memory cards. Looks like the D750 writes at around 70MB/S max (pretty similar to other Nikon cameras that use those cards).

https://www.cameramemoryspeed.com/nikon-d750/fastest-sd-card-speed-tests/

The whole writing jpegs and RAWs in certain orders to cards confuses me. So many different orders it’s confusing what’s best.
 
That's the site I used to choose the card in the first place, it's the Kingston in second place. It definitely helps with emptying the buffer as my other card which has a decent write speed is painfully slow but it's still a small buffer.
 
Back
Top