D500 General Discussion Thread

Michael

I handled a sample D500 with pre-release firmware 0.20 today. The Nikon staff would not allow an image to be taken, but they did let me hand them my SD memory cards, and they inserted them and showed me the "rNUM" value which is the camera's claim of number of frames remaining before buffer saturation.

I had previously set the camera for 14-bit lossless compressed RAW, with most options on (e.g., Active D-Lighting, Auto Distortion Control, VR, Auto ISO, long exposure NR, high ISO NR).

I handed them a very slow, class 10 SD card (previously my measurements indicated that it writes at about 15 MB/s, or less than 1 fps). The LCD on the top showed "r29". They gave me back my card and I handed them a UHS-II card claiming to have a 240 MB/s write speed (I cannot yet test that, as I have no UHS-II devices, but since UHS-I cards claiming 90 MB/s write speeds really write at 72 MB/s, I assume it writes at 192 MB/s or better, or 7 fps or better). Again, the LCD indicated r29. Nikon tends to be conservative about these types of numbers, hence r29 probably means there is room for 30 frames in the internal RAM.

Thus at present, I am hopeful that we're looking at enough RAM in the D500 to hold 30 frames of 14-bit lossless NEFs regardless of card type or speed. That suggests that slow SD cards will have a 3 second buffer at approximately 10 fps, that UHS-I ("up to" 90 MB/s write speed) cards will have perhaps a 4 second or 40 frame buffer (I'm rounding greatly), and that UHS-II ("up to" 240 MB/s write speed) cards will have a 10 second or 100 frame buffer.

My only hesitation/concern is that I saw a video showing a D500 being tested with the same UHS-II card. It looks like there is a "12" on the top LCD, and it only achieved 74 frames in a burst test. That was probably testing 14-bit compressed (not lossless) frames, which are larger, thus reducing the buffer, but 12 versus 29 seems odd. There is also a youtube video with unknown settings, but appears to be NEF based on the top LCD, showing r16. So the mystery remains, but I clearly saw r29 on the camera today.

On other fronts, I believe some people are wondering about the number of AF-Fine Tuning values/lenses that can be stored in the camera's memory. The answer on the sample D500 was 20.

Additionally, some are hoping that zoom lenses will allow storing 2 values (wide-end and telephoto-end). Sorry, but there was nothing to indicate that would be the case. I tried the auto fine tune system twice (once wide, once tele) and only one value was stored in the AF Fine Tuning memory.

For those that care about feel, I personally liked the grip (shape, depth). Not perfect, but superior to the D7100 and D7200 (I have large hands).

I don't have enough experience or tests to say too much about the noise/volume level of the shutter, but I think it sounded comparable to the D7200 which is quieter than the D7100. I think the quiet mode was quieter than the normal modes, but not by enough to bother using it (I feel the same way about the D7100, quite mode is quieter, but not enough to bother with).

I have never had a camera with custom settings banks, but my quick test with them suggested they work like in other Nikon cameras that have them, that is, they do NOT operate like U1/U2 modes (they remember changes you are making during shooting, rather than storing defaults).

I AM: Eagerly Awaiting the Arrival of the D500

Ironheart

D510 will have all the fixes I want but didn't know I needed them...

framer

Ah, the optimists are here...

starralazn

I handled a sample D500 with pre-release firmware 0.20 today. The Nikon staff would not allow an image to be taken, but they did let me hand them my SD memory cards, and they inserted them and showed me the "rNUM" value which is the camera's claim of number of frames remaining before buffer saturation.

I had previously set the camera for 14-bit lossless compressed RAW, with most options on (e.g., Active D-Lighting, Auto Distortion Control, VR, Auto ISO, long exposure NR, high ISO NR).

I handed them a very slow, class 10 SD card (previously my measurements indicated that it writes at about 15 MB/s, or less than 1 fps). The LCD on the top showed "r29". They gave me back my card and I handed them a UHS-II card claiming to have a 240 MB/s write speed (I cannot yet test that, as I have no UHS-II devices, but since UHS-I cards claiming 90 MB/s write speeds really write at 72 MB/s, I assume it writes at 192 MB/s or better, or 7 fps or better). Again, the LCD indicated r29. Nikon tends to be conservative about these types of numbers, hence r29 probably means there is room for 30 frames in the internal RAM.

Thus at present, I am hopeful that we're looking at enough RAM in the D500 to hold 30 frames of 14-bit lossless NEFs regardless of card type or speed. That suggests that slow SD cards will have a 3 second buffer at approximately 10 fps, that UHS-I ("up to" 90 MB/s write speed) cards will have perhaps a 4 second or 40 frame buffer (I'm rounding greatly), and that UHS-II ("up to" 240 MB/s write speed) cards will have a 10 second or 100 frame buffer.

My only hesitation/concern is that I saw a video showing a D500 being tested with the same UHS-II card. It looks like there is a "12" on the top LCD, and it only achieved 74 frames in a burst test. That was probably testing 14-bit compressed (not lossless) frames, which are larger, thus reducing the buffer, but 12 versus 29 seems odd. There is also a youtube video with unknown settings, but appears to be NEF based on the top LCD, showing r16. So the mystery remains, but I clearly saw r29 on the camera today.

On other fronts, I believe some people are wondering about the number of AF-Fine Tuning values/lenses that can be stored in the camera's memory. The answer on the sample D500 was 20.

Additionally, some are hoping that zoom lenses will allow storing 2 values (wide-end and telephoto-end). Sorry, but there was nothing to indicate that would be the case. I tried the auto fine tune system twice (once wide, once tele) and only one value was stored in the AF Fine Tuning memory.

For those that care about feel, I personally liked the grip (shape, depth). Not perfect, but superior to the D7100 and D7200 (I have large hands).

I don't have enough experience or tests to say too much about the noise/volume level of the shutter, but I think it sounded comparable to the D7200 which is quieter than the D7100. I think the quiet mode was quieter than the normal modes, but not by enough to bother using it (I feel the same way about the D7100, quite mode is quieter, but not enough to bother with).

I have never had a camera with custom settings banks, but my quick test with them suggested they work like in other Nikon cameras that have them, that is, they do NOT operate like U1/U2 modes (they remember changes you are making during shooting, rather than storing defaults).

I AM: Eagerly Awaiting the Arrival of the D500

thanks for the observations. glad you like the D500.
question, what would the internal flash memory have to do with the type of external memory used? the difference in speed of a card should only affect how long it takes for that 29 file buffer to clear. the only thing that should affect that 29 number is the settings in camera..

Ironheart

I wouldn't put too much stock in the buffer on the demo units. Nikon claims this in all of their literature on the D500:
NEF (RAW), Lossless compressed, 14-bit: 200 shots
NEF (RAW), Uncompressed, 14-bit: 79 shots
No qualifications on what type of card, speed, etc... That means they are going to put one beefy buffer in this puppy.

https://cdn-4.nikon-cdn.com/e/Q5NM96RZZo-YRYNeYvAi9beHK4x3L-8go_p7JUL6JpQM9h_9xTbwyw==/PDF/D500_Comparison_Sheet.pdf

Ironheart

In this doc:
http://chsvimg.nikon.com/lineup/microsite/d500/common/pdf/technology-digest.pdf

They do list the caveat:

With Lexar Professional 2933x XQD 2.0 (64 GB).

I guess time will tell

Michael

starralazn: A fast card allows frames to be removed from the camera's internal RAM before the RAM fills up, thus a fast card allows a deeper buffer since the practical buffer depth is (RAM divided by filesize) plus (number of frames written to card before the buffer fills).

As Ironheart pointed out, a lot of the market hype carries the caveat of testing with the 64GB Lexar Professional 2933x XQD 2.0. That card claims write speeds up to 400 MB/s (same for the 128 GB card, with the 32GB card slightly slower at 390 MB/s). Since a 14-bit lossless 21 megapixel file will probably be around 27 MB, we only need cards that can write at 270 MB/s (which marketing departments might advertise as 340 MB/s).

Of course, there is no way to know if the demo units even have the same amount of RAM as the production models. Indeed, only time will tell.

spraynpray

@Ironheart: I see that comparison says that the D7200 does not support powered aperture control - is that right? I thought all models post D750/D7100 (incl) had that. If not, does that mean the newer E lenses are not compatible with my D7100?

Either I misunnerstan or it's a typo surely...

Msmoto

Why a D500 when I have a D4......? Well, that is an excellent question. The reasons are to have a bit of a longer look, about the same FPS, yet record the image in DX size with 20.9 MP vs. 6.8 MP as the cropped D4 gives. Also, I suspect the focus is better, and with the more data, post processing/editing will give nicer results.

The ISO may be an advantage as well, but the D4 has had adequate ISO in nearly every situation I have attempted.

The one piece of information I do not have is the buffer capacity for continuous shooting.....how many images will it hold shooting 14 bit lossless compressed?

Paperman

Ironheart said:

I wouldn't put too much stock in the buffer on the demo units. Nikon claims this in all of their literature on the D500:
NEF (RAW), Lossless compressed, 14-bit: 200 shots
NEF (RAW), Uncompressed, 14-bit: 79 shots
No qualifications on what type of card, speed, etc... That means they are going to put one beefy buffer in this puppy.

Such a large buffer eliminates the need for ultra fast cards. I wonder how many will be aware of that and opt out of them ( I saw an SD card for CAD$999 last month ! )

Ironheart

spraynpray said:

@Ironheart: I see that comparison says that the D7200 does not support powered aperture control - is that right? I thought all models post D750/D7100 (incl) had that. If not, does that mean the newer E lenses are not compatible with my D7100?

Either I misunnerstan or it's a typo surely...

The E lens I have (16-80mm f/2.8-4) works great on my D7200. Could be a typo. Which doc are you looking at?

Edit: The D500 comparison doc is correct. The Power Aperture feature allows you to smoothly open or close the aperture during movie shooting, so that you can adjust the exposure on the fly or change the depth of field, without it being noticeable in the recording. This requires a second actuator motor that Nikon first started using in the D800 I believe. Otherwise it has to cycle the shutter to change the aperture. This is one of the few differences between the D750 and the D7200. The E series lenses are the exception to this rule since they effectively have their own actuator. So using my 16-80 I can change aperture during movie shooting.

spraynpray

Ironheart said:

spraynpray said:

@Ironheart: I see that comparison says that the D7200 does not support powered aperture control - is that right? I thought all models post D750/D7100 (incl) had that. If not, does that mean the newer E lenses are not compatible with my D7100?

Either I misunnerstan or it's a typo surely...

The E lens I have (16-80mm f/2.8-4) works great on my D7200. Could be a typo. Which doc are you looking at?

The one you linked to above:

https://cdn-4.nikon-cdn.com/e/Q5NM96RZZo-YRYNeYvAi9beHK4x3L-8go_p7JUL6JpQM9h_9xTbwyw==/PDF/D500_Comparison_Sheet.pdf

Ironheart

Refresh to see my edit above. Pre-coffee (:|

fiziks

Msmoto said:

Why a D500 when I have a D4......? Well, that is an excellent question. The reasons are to have a bit of a longer look, about the same FPS, yet record the image in DX size with 20.9 MP vs. 6.8 MP as the cropped D4 gives. Also, I suspect the focus is better, and with the more data, post processing/editing will give nicer results.

The ISO may be an advantage as well, but the D4 has had adequate ISO in nearly every situation I have attempted.

The one piece of information I do not have is the buffer capacity for continuous shooting.....how many images will it hold shooting 14 bit lossless compressed?

I talked to a local pro a couple weeks ago. He was shooting a D3s and noticed me shooting my D300s. He asked if I'd heard about the D500. I said that I was thinking of getting one and asked if he was going to go for a D5. He said he had a D4s, but preferred the D3s and would probably trade the D4s (or both?) in for a pair of D500s. That kinda blew me away. Unfortunately, that was all the conversation he had time for.

starralazn

@fiziks, i think you could probably still get a D4s for at least 4000 (or more to account for taxes) so that sounds about right to me (if he really didnt like his D4s...)

Michael

I just found a Lexar (Japan) video on youtube demonstrating their memory cards on a D500. They do some 10 second burst tests to see how many frames can be taken in 10 seconds ( ). The marketing strategy is to get people to buy their faster cards (the video compares their fastest cards with some of their other cards, both in frames taken in 10 seconds and in how long it takes to write those frames to the cards).

The SD card test starts around 1:17. I can't tell what type of compression setting is being used, but it is a 14-bit NEF test. In the video, "r29" is clearly visible on the LCDs, further confirming that we'll have at least a 3 second buffer at 10 fps, regardless of card speed.

Their UHS-II card claims to write at up to 260 MB/s (but that figure is probably calculated by their marketing department). Based on the video, it looks like the buffer saturated at 39 frames, much lower than I would have expected. The other video I saw (but can no longer find) was with a Toshiba card claiming write speeds up to 240 MB/s, that saturated at 74 frames.

The mystery lives on.

Paperman

Math does not work for Toshiba's 45 frames ( 74-29 ) at 240 MB/s clearance - at 30-40 MB per photo.

manhattanboy

@Michael Thank you very much for that video. The two takeaways for me are if you buy the D500 you would be foolish not to get the latest XQD cards, and two, the old XQD cards still blow away the fastest/newest SD cards.

Ironheart

@manhattanboy If I could upvote your post I would. Why buy a Ferrari and cheap out on the gas.

fiziks

Keep in mind that the SD writer in the camera may be the limiting factor. One would hope that they put the latest and greatest in the D500, but they may skimp on the max SD write speed under the assumption that anyone who really wants a long burst will use XQD.

Ironheart

I already bought my Lexar high-test gas. I'm ready

Sports

I, too, couldn't resist, and I now have one of those 2933x Lexars sitting and waiting on my desk. "Only" 32 Gb, though.
When the prices fall, I'll get a 128 Gb or something.

Paperman

manhattanboy said:

@Michael Thank you very much for that video. The two takeaways for me are if you buy the D500 you would be foolish not to get the latest XQD cards, and two, the old XQD cards still blow away the fastest/newest SD cards.

Can anyone explain to me the need for ultra fast cards if the buffer is 79/200 RAWS as in specs ? - assuming of course, the 29 RAW buffer mentioned is wrong . It may be actually be foolish to pay hundreds of dollars when $20-40 will do the same job.

BVS

Paperman said:

Can anyone explain to me the need for ultra fast cards if the buffer is 79/200 RAWS as in specs ? - assuming of course, the 29 RAW buffer mentioned is wrong . It may be actually be foolish to pay hundreds of dollars when $20-40 will do the same job.

It's only 79/200 when using an ultra fast card. With an ultra fast card the camera is able to write the photos to the card and clear the buffer almost as fast as new photos are being taken, thus it can go for a long time (79/200) before it fills up. It's kind of like trying to fill up a bathtub with the drain open - it takes a lot longer the faster the water is draining out. With a slower card the water/photos don't drain out as fast and the buffer fills more quickly. In theory, if the card was fast enough the 'buffer' could be unlimited.

That said, I wonder if you could get an effectively unlimited buffer by dropping down the CH speed a notch or two, say 8fps or so?

heartyfisher

BVS said:

.....

That said, I wonder if you could get an effectively unlimited buffer by dropping down the CH speed a notch or two, say 8fps or so?

I am sure you are right there .. if the card is fast enough to clear at 8 FPS. I am bad at maths and its too much brain capacity for me to calculate the minimum FPS if the buffer saturates at 40,50,70 etc ... maybe some math geek here can do it for us :-) ie it would be nice to know that for a certain card if the buffer fills up at a specific value, switching to a certain FPS should get you to say 100 or 200 frames before the buffer fills.

Michael

If a card can accept 7 frames per second (which I am hopeful a fast UHS-II card will be able to do), then presumably the buffer would be 200 when shooting at 8 fps or less.

For a fast UHS-I card that can accept 2.7 frames per second, you'd have to go down to 4 fps just to get a 20 or 21 second buffer depth (84 frames). Hence UHS-I is now offically slow. For serious burst usage, we'll need UHS-II or XQD, or stick with the built-in RAM which will probably provide a 30 frame buffer.