Reactions from Nikon D5/D500 Launch Event

[ShunCheung]

<p>Nikon USA has started a series of D5 and D500 launch events this week: http://www.nikonusa.com/en/learn-and-explore/events/d5-d500/index.page<br>

For example, B&H had two sessions this past Monday, Adorama held its event yesterday (Tuesday) and there were others (e.g. Santa Barbara). There will be some more in the coming days. I'll be attending one shortly myself.</p>

<p>For those who have been to the event, what are your impressions?</p>

<p>Also, for those who are not attending but have questions, perhaps we/I can help ask. For example, I am interested in what the real buffer size is, especially on the D500. The capability to copy from one memory card to another can be important, especially on the D500. Since the D500 uses one XQD card and one SD card, and most laptops have an SD card slot built int, as long as we can copy from XQD to SD, we can go through an SD card to backup images on the XQD without using a card reader (or even without a PC if you just backup to cheaper SD cards).</p>

[Dieter Schaefer]

<p>I went to the event at the local camera store last night. You get to see all the promotional videos that are already available on the web. Then you get to play with the cameras ;-)</p>

<p>My main interest was the D500 - and the first disappointment was that the samples available are clearly pre-production prototypes. There were three D500 bodies - they showed "remaining buffer" values between "r19 and r29" - none of the three went for 200 frames despite the XQD card in the (taped) card slot. The firmware on those samples clearly isn't in its final form. No such issue on the D5 - it fired along at 12fps for 200 shots before stopping. If the price tag and the 3-pound weight of Nikon's flagship don't bother you, the camera definitely appears to be a gem.</p>

<p>Back to the D500 - it feels like a bigger D750 - almost the same grip shape and the same chassis construction (magnesium back, carbon fiber front). Weight is almost identical. It doesn't have the narrower top LCD of the D750 though and not the slimmed-down body either. The viewfinder is definitely "FX-like" with its 1.0x magnification (the D750 is 0.7, so there is not much difference in appearance). The control layout continues along the lines of recent Nikon DSLRs - favoring two-handed operation (button press with the left hand, dial turning with the right). I immediately liked the AF-selector joy-stick in the back - a joy to use ;-) The dedicated AF-On button is right there next to it - perfect placement. AF coverage in the viewfinder is excellent - right up to the (short) edge. We were told that the AF module of the D5 was put into the D500 without modification - hence the much larger covered area compared to other DX cameras.</p>

<p>The Expeed5 processor promises a 30% increase in speed - and moreover is frugal with the batteries - expect to squeeze a few more shots out of the EN-EL15. On the D5, Nikon managed in a real-life shooting scenario some 8000 shots from a single battery.</p>

<p>High ISO performance of the D500 was naturally another item I was interested in - but there is the already mentioned caveat of the camera being a pre-production sample with pre-release firmware. Also, one can only evaluate on the back LCD screen (tiltable, BTW, but feels solid). The range extends to 51600 - personally I wouldn't use that or 25600 - and 12800 only in a pinch. 6400 looked usable, 3200 definitely so, and 1600 fairly clean. Please note that tolerance to image noise varies from person to person and also with use - for me, the D300, D7100, and D810 have the same noise level when looked at on the screen in 100% at what I consider to be their maximum usable ISO of 1600. For the D700, that limit for me is ISO 3200 - and I would usually apply some amount of noise reduction on those images. The D500 appears a bit above the level of the first three, and a bit below that of the D700. Again, this impression may change with a production model and certainly when one can process the RAW files and view the images on a good monitor.</p>

<p>AF acquisition was sure and fast with the D500 (used the 300/4 PF and 16-80/2.8-4) in the rather dimly lit store. Not much chance to evaluate tracking performance inside the store though. We were told the D500 is the first Nikon DX camera (and the D5 the first FX) with a separate processor for AF - previously, the Expeed processor had to do that work too. Real-life test will show where the differences to the AF in the older cameras really lies.</p>

<p>We get the usual "moved the cheese" we all grew accustomed to by now - the mode button is now on the left (and I did not see an option to reprogram the ISO button on the right to be the mode button). So if one plans on using the D500 alongside any other Nikon body, then one needs to pay attention to yet another change in the control layout.</p>

<p>Didn't pay any attention to the video features. Didn't play with the touch screen features of the D500 (and the limited ones on the D5 (in playback mode as well as when entering text). Didn't play with the D500 snapbridge - but it is intriguing. Apparently, even firmware updates to the camera can now be delivered via your smartphone. So don't be surprised when you turn on the D500 and there is a message on the screen alerting you to a firmware update? Hope we get spared to ever see the message "you got hacked!"</p>

<p>And no, I didn't pre-order one. I am definitely sufficiently intrigued to want one - but I want to be able to test a production sample first (and read a few reviews from trusted sources) before spending what will surely amount to $3K (body + battery grip + XQD card(s)). To me, the appeal are the AF (performance yet to be fully determined) and the fast speed. For more general type shooting than what I do, I am not sure I would pick the D500 over the (much cheaper) D7200 (if staying with DX) or over the (equally expensive) D750 (if moving to FX is an option).</p>

[ShunCheung]

<p>Concerning battery usage, the CIPA standard measurement is that for a camera with a pop-up flash, battery usage is measured with the flash going off in every other capture. Since the D500 has no pop-up flash, its CIPA figure will appear to have much better battery life than previous DX bodies such as the D7100, D7200, and D300, etc. In reality, the CIPA figures are comparing quite different situations.</p>

<p>Prior to the D500, the only Nikon DX DSLR bodies without a pop-up flash were the D1 and D2 families.</p>

[Dieter Schaefer]

<blockquote>

<p>In reality, the CIPA figures are comparing quite different situations.</p>

</blockquote>

<p>Understood - I was relaying real-life information given by the presenters. CIPA value for the D5 is 3780 - the real-life number we were given was 8K+. </p>

[Richard Williams]

<blockquote>

<p>We were told that the AF module of the D5 was put into the D500 without modification - hence the much larger covered area compared to other DX cameras.</p>

</blockquote>

<p>Sounds great for DX, but slightly less great for FX (though of course better than now). Did you get a chance to assess how this worked for the D5?</p>

<blockquote>

<p>So if one plans on using the D500 alongside any other Nikon body, then one needs to pay attention to yet another change in the control layout.</p>

</blockquote>

<p>I <strong>really</strong> wish they'd stop doing this!</p>

[Andrew Garrard]

<blockquote>I <b>really</b> wish they'd stop doing this!</blockquote>

 

<p>I <b>really</b> wish they'd make everything fully programmable (preferably with chording for a few more options) so we could sort out the mess ourselves. I liked ISO on the right (mostly I don't care about it being aliased on the record button), but I'm not that keen on mode having moved. With the D810, I wasn't very pleased that the exposure mode button moved. With the D800, I wasn't very pleased that the AF mode selection moved (compared with the D700). The only reason I can see not to be fully programmable is the risk of making some long menus slightly longer (but more consistent) and the confusion of technical support - and they can always talk people through resetting the camera to default settings before they go any further.</p>

[Dieter Schaefer]

<blockquote>

<p>Did you get a chance to assess how this worked for the D5?</p>

</blockquote>

<p>Coverage for the D5 is slightly larger than for the D4 - the outermost rows on the D5 are just outside the limits of the D4 coverage. In addition, the outer rows have five vs previously 3 AF spots - so there is a bit more in the corners as well. Here's an image that shows the difference: <a href="http://2.static.img-dpreview.com/files/p/TS560x560~forums/57039123/5920a0f6a1db48f7aa379287dfc6c0f8">http://2.static.img-dpreview.com/files/p/TS560x560~forums/57039123/5920a0f6a1db48f7aa379287dfc6c0f8</a>.</p>

<blockquote>

<p>I <strong>really</strong> wish they'd stop doing this!</p>

</blockquote>

<p>And you aren't alone on this. But obviously those who have Nikon's ear on this have a different take on where things belong. Looking at the top deck of the D500, I'd dispense with the video recording button (could be easily a fourth setting on the ring around the shutter release or actually programmed as a different mode and released with the shutter release (what a novel concept)). That would make space for a mode button on the top deck, together with the EV comp and ISO button. Then move the BKT to the other side of the prism housing. Right next to it would be the metering mode. Then move the AF selector to the other side of the mount housing as well. Voila, everything important (to me) is on the right side. And there's still two function buttons that could easily be programmed to hold QUAL and WB functions (which I rarely change and hence don't mind in their current locations - other may feel differently). Now, there's only one more change - the different drive modes need to come off that dial and onto a button/command dial scenario - of course for the right hand to deal with.</p>

<p>But as I have been repeatedly told - Nikon doesn't build a camera for me. I do wonder though who is calling those control placement shots?<br>

<br /> FWIW, the back of the D5 looks identical to the one from the D4 - the "cheese moving" is restricted to the top left buttons and the right hand side of the camera.</p>

<p>Menu: there's a new layout for the menu to assign functions to the various buttons on the D500 (didn't check the D5) - couldn't figure it out in the time I had - but it's radically different from what we currently have.</p>

[dan_brown4]

<p>D5 is three pounds! Yikes.</p>

[ShunCheung]

<p>Actually I think the Multi-CAM 20000 provides too much coverage on the DX-format D500. Some of the cross-type AF points are very close to the edge on the short (24mm) sides of the frame. One rarely has a subject placed that close to the edge. I kind of wish on the D500, the AF points are more concentrated to the center. Yeah, I know that I am picky and am hard to please.</p>

<p>That AF module works better on FX, but it is probably designed for FX to begin with.</p>

[Dieter Schaefer]

<blockquote>

<p>One rarely has a subject placed that close to the edge.</p>

</blockquote>

<p>Yeah, it would probably have been sufficient to fill the 1.3x crop frame (i.e. eliminate the outer row of AF points). But heck, if I am using one point from the that 2nd row, and in D25 coverage mode (that's the smallest number to select, there no longer is a D9), the outer row got some work to do too. </p>

[eric_arnold]

<p>the d500's ISO performance will likely determine whether i get one or not. i would need a clean 3200 and usable 6400. would you say it achieves that?</p>

[Dieter Schaefer]

<blockquote>

<p>i would need a clean 3200 and usable 6400. would you say it achieves that?</p>

</blockquote>

<p>Given that you are using a D3S as a base for comparison, I am inclined to answer no. What I've seen on the back LCD screen of the D500 when zooming in would have made me apply some NR in post. Question is though whether the finalized firmware will do better - and the quality of the RAW files is still an unknown.</p>

[ShunCheung]

<p>I went to the launch event at Keeble and Shuchat in Palo Alto on Wednesday night. Mr. Shuchat himself was on hand to introduce the Nikon representatives. It was, as expected, an hour-long infomercial with some question and answer session and then a few short videos on how great the D5 and D500 are, and there are several comparisons against the competition, namely the Canon 1DX and 7D Mark II, respectively.</p>

<p>That event was very crowded with around 100 participants. Unfortunately they only had one D5 and one D500 sample on hand. I had to wait quite a while to hold each camera, briefly. There were a number of detailed questions even the Nikon reps didn't know the answers off hand. They did confirm that they expect 90% of the D5 shipped will use XQD cards and they highly encourage that configuration. They will be able to swap the CF module with the XQD one later on, but they don't know the cost yet.</p>

<p>They also confirmed that the D5 will ship towards the end of March, but they wouldn't give any date for the D500. Nikon USA had given me (some time in) April as the new date. Nikon Japan has second half of April on their web site.</p>

<p>I did check the real buffer on the D500. Initially the D500 showed r99, but it was set to JPEG capture. When I switched to RAW, it showed r12. Therefore, the D500's buffer indeed seems shallow and it highly depends on dumping the images onto a fast memory card. I think the D5 shows r53.</p>

[Dieter Schaefer]

<blockquote>

<p>That event was very crowded with around 100 participants. Unfortunately they only had one D5 and one D500 sample on hand.</p>

</blockquote>

<p>That's bad planning on Nikon's part. Here in Santa Barbara less than 50 attended - and there were three D500 bodies (actually 4 - but one was broken due to an attendee dropping it the day before). Three D5 bodies as well.</p>

<blockquote>

<p>I think the D5 shows r53</p>

</blockquote>

<p>The ones I handled all showed r99 - didn't check the QUAL setting though. Shun had linked previously to this thread on dpreview: <a href="http://www.dpreview.com/forums/post/57053571">http://www.dpreview.com/forums/post/57053571</a>. Apparently, the D500 displayed at CES had a later (production?) firmware since the D500 and achieved the quoted 200 frames before slowing down - none of the three pre-production D500 I handled at the event got anywhere close to that number (with XQD card installed). The poster in that thread saw r14 for all NEF modes except 12-bit (which showed r16).</p>

[Dieter Schaefer]

<p>So the D5 has a deep buffer and the D500 does not - yet with the XQD card both achieve the same 200 frame burst without slowing down. Begs the question - why does the D5 need so much memory when the D500 can do the same thing with a lot less?</p>

[ilkka_nissila]

If you use a slower card you can do longer bursts if the "real" buffer is large (than if it were small). It would seem the D500 is more

dependent on fast cards than the D5, in extended fps shooting (however the final cameras may be different).

[jon_porter]

<p>I, too, was at the Keeble & Shuchat event. As Shun said, it was disappointing there were only two cameras for quite a large crowd to handle. Plus the D5 had the 14-24mm lens attached – not the best choice when demonstrating a new camera. But the D500 had the 16-80mm, a very impressive combination. My take-away from the evening was that Nikon is committed to XQD. Both reps commented several times that the performance of the cameras was designed around the new cards.</p>

[Dieter Schaefer]

<blockquote>

<p>If you use a slower card you can do longer bursts if the "real" buffer is large (than if it were small).</p>

</blockquote>

<p>Forgot about the small percentage of D5 cameras that will be sold with two CF card slots instead of two XQD cards. For the latter, the buffer size should be quite immaterial - or are there "slow" XQD cards?</p>

[ilkka_nissila]

Well the first generation of XQD cards were not all that fast (128MB/s read, 80MB/s write). Some D4 users may have

them around and want to use them on the D5. The 200 NEF burst at 12 fps probably is achieved using the latest XQD

cards (400MB/s read 350 MB/s write). And yes, some D5 users will buy their camera or have it modified with the CF

interface to use their old cards. And for maximum performance with those, a large physical buffer memory is a benefit.

[ShunCheung]

<p>When you half press the shutter release on the Nikon DSLRs, the viewfinder will show the number of frames remaining in the (real) buffer, as some rNN numbers (number of frames remaining). If you shoot a burst and then immediately look into the viewfinder, you'll see that number increases gradually as the camera dumps the images from the buffer onto the card, thus freeing up more space.</p>

<p>The sample cameras Nikon had last night at Keeble and Shuchat were set to JPEG capture when I handled them, and the rNN numbers were high. Therefore, I quickly switched them to RAW, but I didn't have time to specify 12 vs. 14 bit and compression. A quick glance validated that the D500's real buffer is shallow, while the D5 has a real buffer that will last you a few seconds at 12 fps, hence you maybe able to get away with a slower XQD card as long as you don't shoot long bursts.</p>

<p>Some of the early XQD cards are not all that fast, as we discussed on this recent thread: http://www.photo.net/nikon-camera-forum/00dgnq<br /> Nikon is clearly pushing XQD as the way to go. Yesterday I mentioned that XQD prices will likely come down soon, and the Nikon rep couldn't agree more. I think the D500 may forces SanDisk to get into the XQD game as that is more a mass-market camera.</p>

<p>The Nikon rep also mentioned that she got a 1G CF card way back in 1999 along with the original D1, which I first saw at Photo+Expo in New York City back in 1999. She said that 1G CF cast cost over $1000 back then. She can't even give that card away now.</p>

<p><a href="/photodb/user?user_id=502764">Jon Porter</a>, totally agree that it was not a good idea for Nikon to put a heavy, especially front heavy 14-24mm/f2.8 AF-S on the demo D5. That lens made the camera really heavy and awkward to handle.</p>

<p>I also found out that even the Nikon reps were not that familiar with the D5 and D500 yet. There were a number of questions that they couldn't answer. For example, I went with a friend who asked them which 15 AF points are compatible with max f8 lenses, and both reps could not answer. After the event, I quickly found out that on page 13 of the D5 brochure, there is a diagram showing those 15 AF points (some of them are not user selectable). They are in a + shape in the center of the frame.</p>

<p>You can find the D5 brochure here:<br /> http://cdn-4.nikon-cdn.com/e/Q5NM96RZZo-YRYNeYvAi9beHK4x3L-8go_p7JUL6JpQMwSj_xzTyyQ==/PDF/D5_Brochure.pdf</p>

<p>Another interesting change Nikon has made is moving the MODE button to the left side of the top, and now the ISO button is right behind the shutter release, as shown on the top view of the D5: http://static.bhphotovideo.com/explora/explora/sites/default/files/Nikon-D5-Top.jpg<br>

Nikon thinks it would be easier to change the ISO quickly this way. The D500 is like that also.</p>

[Dieter Schaefer]

<blockquote>

<p>Another interesting change Nikon has made is moving the MODE button to the left side of the top, and now the ISO button is right behind the shutter release, as shown on the top view of the D5</p>

</blockquote>

<p>I already mentioned that one above - and to me it is the worst change being made. I always resented the mode dial of the D7100 because of it location on the left - but at least it requires only one hand to operate. The mode dial on the left forces a two hand operation on me - so with a heavy lens, I have to find a place to set everything down.<br>

From a review video I just watched, I got the info (which wasn't available when I was at the launch event - I asked) that the video recording button can be re-programmed to act as the mode button. I hope that turns out to be true! Now there only needs to be a way to duplicate the AF area mode selection with some button that's reachable with the right hand. </p>

[eric_arnold]

<blockquote>

<p>Given that you are using a D3S as a base for comparison, I am inclined to answer no.</p>

</blockquote>

<p>the D3s did set a high bar for ISO performance, but that was 2009 technology. i dont expect an APS-C sensor to leapfrog to equivalence with an FX sensor, but i would think a reasonable bump in performance from the previous DX generation, which would be the d7200, should be expected. especially because they kept the mp count at 20 specifically for noise handling. so, i'm gonna remain cautiously optimistic for now. </p>

[Dieter Schaefer]

<blockquote>

<p>i dont expect an APS-C sensor to leapfrog to equivalence with an FX sensor</p>

</blockquote>

<p>The way the ISO advantages of FX over DX and the DX evolution are thrown around over the years - 1 stops here, 1 1/2 stops there, and another 1 1/2 thrown in for good measure - one would think that the current generation DX sensor has eclipsed the first generation FX one (D3/D700) by a wide margin by now. At least from my point of view, a D7100 does not reach the noise level of a D700 - dxomark results confirm that the D7200 has closed the gap halfway between the D300 and the D700 (and the D7200 results only marginally better than the D7100) - or about 1 stop above the D300 and still 1 stop below the D700 - 2 stops has always been the most the difference between the D300 and the D700 was considered to be (pretty much what the dxomark measurements show - it's actually closer to 1 1/2 stops). These observations are based on the normalized 8MP output results - on the pixel level, the D7200 has gained nothing on the D300 at all - the Screen SNR18% results for the D300 and D7200 superimpose almost exactly.<br>

So the D500 would need to show a 1 stop over the D7200 - rather doubtful when one considers that it took three generations to get one stop above the D300 level.</p>

<blockquote>

<p>the D3s did set a high bar for ISO performance, but that was 2009 technology</p>

</blockquote>

<p>On the pixel level, it's still the champ in Nikon's camp (and has to yield by a very small margin to the Sony A7S/SII). Again, that's for the Screen dxomark results - not the more commonly cited Print results that favor higher MP sensors (because of the noise reduction that comes from reducing everything to a 8MP output).</p>

[ilkka_nissila]

<p><em>1 stops here, 1 1/2 stops there, and another 1 1/2 thrown in for good measure</em></p>

<p>People tend to get excited when a new camera is announced. In reality there hasn't been much change in light collection efficiency after the D3s and there hasn't been much change in dynamic range since D7000; these are the breakthrough cameras in terms of sensor design. There has been some subtle improvement since but it is really subtle. The manufacturers would like us to believe that the rapid pace of image quality improvements is continuing in new releases (by increasing the nominal ISO in a ridiculous way) because they realize that people only upgrade if they can expect better results by doing so. But buyers are expected to be smart and see through such fluff. Buy a D500 for its new AF, high fps, and large number of continuous shots in a burst, not for any expectation of better image quality.</p>

<p><em>on the pixel level, the D7200 has gained nothing on the D300 at all - the Screen SNR18% results for the D300 and D7200 superimpose almost exactly.</em></p>

<p>First of all that's like saying that a half a gallon of milk today only has the same total amount of nutrients than a gallon of milk a few years ago, and complaining that it's not more. Midtone (18%) SNR is primarily determined by the number of photons detected during the exposure and that hasn't changed all that much (per sensor area) since the D3s; that camera is already incredibly efficient in collecting photons for detection. The 2007 D300 was not as efficient in collecting photons hence today a smaller area of the D7200 sensor produces as much tonal content than a twice as large area on the D300 sensor, a remarkable achievement in itself, but this evolution is approaching its end due to physical limits. In the shadows the improvement has been much more significant and if you click "dynamic range" on dxomark you'll see that even at the pixel level the dynamic range is much greater on the D7200, a testament to its lower (read) noise sensor.</p>

 

[ShunCheung]

Maybe there will be some initial bugs to work out, but I am quite sure that the D5 and D500 will turn out to be excellent.

However, there is no need to buy into all the marketing hype. For example the D7200 is already very good. The D500 is

not going to be a huge jump forward.

 

I am wondering whether anyone else besides Dieter, Jon, and me has attended these events? There are quite a few of

them this week. I read that some guy in the Seattle area went in three consecutive days. I could have done something

similar, but given that they are so full, it would have been unfair not to give others a chance. And it is mostly an

infomercial anyway. Hearing it once is enough.

 

Eric, did you get a chance to go? There was one at Berkeley yesterday and one in San Francisco tonight (already full).

 

BTW, our Nikon rep mentioned that there are some 80K D500 preorder in the U.S. already. Whether people have ordered

from multiple stores and get whatever comes first is unknown. However, he also emphases that If you preorder at the

event, you would receive your camera as soon as Nikon starts shipping.