Author Topic: Nikon D5 - first impressions  (Read 146404 times)

Bjørn Rørslett

  • Fierce Bear of the North
  • Administrator
  • ***
  • Posts: 8252
  • Oslo, Norway
Re: Nikon D5 - first impressions
« Reply #150 on: March 28, 2016, 23:11:02 »
Assumption #2 is increasingly becoming the current trend. Numbers are easy to interpret as being "objective" measures.

Andy

  • NG Member
  • *
  • Posts: 314
Re: Nikon D5 - first impressions
« Reply #151 on: March 28, 2016, 23:25:38 »

simsurace

  • NG Member
  • *
  • Posts: 835
Re: Nikon D5 - first impressions
« Reply #152 on: March 28, 2016, 23:40:13 »
The last result is puzzling, but I don't think their method (i.e. Using ACR to push back underexposed shots) is the best way to measure what they want to measure ('dynamic range'). I don't know what they know, but neither do I know what is happening to the raw data as you lift exposure in ACR, and to what extent those manipulations depend on the camera profile. You have to wonder why analyses are not done on the raw data itself rather than on data filtered by a raw converter.
Simone Carlo Surace
suracephoto.com

Ilkka Nissilä

  • NG Supporter
  • **
  • Posts: 1689
  • You ARE NikonGear
Re: Nikon D5 - first impressions
« Reply #153 on: March 28, 2016, 23:57:49 »
I would guess since DXO will do their analysis based on the raw data, dpreview doesn't want to do the exact same thing but something different. They'll no doubt show dxo results once those are published.

I think both DXO and dpreview overemphasize the importance of base ISO dynamic range. The D5 is a fast camera and it is likely that the fast analog-to-digital conversion adds some noise that we wouldn't see in slower cameras. I don't see a huge issue here but this does make the D5 a little bit less general purpose than it might have been. If shooting action in bright sunlight one may wish for as much dynamic range as possible but want to retain the AF and fps of the D5. I don't need high fps often so I can use the D810 for most high contrast scenes and there is, of course, exposure blending. The D5 isn't a landscape camera so it is strange why reviewers test it like it were one. It's a bit like testing studded winter tires in the middle of summer and complaining about lack of friction and loud noise.

It does seem surprising that Nikon were not able to meet the D4s's DR at base ISO ... only 1 fps difference in speed (more if shooting at 14fps of course, without viewfinder). But, I don't think the difference is a big deal.

Evaluating the quality of real world images is subject to the effects of whatever software one is using. This can cloud the differences between hardware. However I think ultimately most people take a practical approach and see what the camera can do in the real world and evaluate that in their own usage.

bjornthun

  • Guest
Re: Nikon D5 - first impressions
« Reply #154 on: March 29, 2016, 03:40:55 »
The last result is puzzling, but I don't think their method (i.e. Using ACR to push back underexposed shots) is the best way to measure what they want to measure ('dynamic range'). I don't know what they know, but neither do I know what is happening to the raw data as you lift exposure in ACR, and to what extent those manipulations depend on the camera profile. You have to wonder why analyses are not done on the raw data itself rather than on data filtered by a raw converter.
Underexposing -1 EV seems to contradict the ETTR rule. Shouldn't they also try to find out how extract the maximum DR they possibly could, and then report what result they got and how they got it, i.e. how did they expose to acquire the result. How do we know that the testing methodology, -1 EV undeexposure, is valid for the D5 or any other camera, or across generations of cameras, D3 - D4 - D5? That's nine years of camera development, a long time in the computer age.

bjornthun

  • Guest
Re: Nikon D5 - first impressions
« Reply #155 on: March 29, 2016, 03:44:53 »
Andy one: The Cheese would have been nicer with the 45PC. Focus on the well lit cheese the somewhat chaotic dark structure as a blurry BG.

Andy two: Did you test some slow AF primes like the 85s and the 50s? How does the AF on the D5 differ compared to earlier AF system incarnations?

Andy three: I was tinking a lot about your concept of the "photon budget". Yes, the D600 has 3x photon budget compared to the D3 ... BUT ...  We compare apples & oranges here as one is measured at "base ISO" 200 the other at 100. Should we not look at the photon budget depending on our intended light levels?

http://sensorgen.info/NikonD600.html
http://sensorgen.info/NikonD700.html
http://sensorgen.info/NikonDf.html

That said it seems like a great idea for Nikon to shell out another D5, name suggestion D5L (for "Low" ISO) featuring the buffer, the fresh Exspeed, the AF, the ruggetized body, all I want and need as a cradle for the 24MP genius.

The combination of the D5 and the D5L as a two body set will give the best of both worlds to any PJ or agency worker.
The so-called "PDR" of the D5 seems to oscillate around the downward trend curve at least until we reach 3200 ISO. Why is that? It seems a bit out of the ordinary, when you compare to other cameras.

simsurace

  • NG Member
  • *
  • Posts: 835
Re: Nikon D5 - first impressions
« Reply #156 on: March 29, 2016, 08:28:34 »
I think both DXO and dpreview overemphasize the importance of base ISO dynamic range.
But where is the supposedly better high ISO performance coming from if not from a lower noise floor? This lower noise floor should also imponge on the base ISO dynamic range. But I guess that the noise floor is relevant in a relative sense, i.e. relative to the full well capacity. So while the full-well capacity might have been reduced, the noise floor is lower in a relative sense. A high full-well capacity is relevant for photon counting statistics, that's important when light levels are high. If light levels are low, i.e. at high ISO, the other noise sources are more important.
Simone Carlo Surace
suracephoto.com

richardHaw

  • Cute Panda from the East...
  • NG Supporter
  • **
  • Posts: 3133
  • Your lens loverboy
    • Classic Nikkor Maintenance and DIY
Re: Nikon D5 - first impressions
« Reply #157 on: March 29, 2016, 08:56:26 »
https://www.youtube.com/watch?v=69U9RuJ5UnM

looks like somebody is not happy with the D5 :o :o :o

simsurace

  • NG Member
  • *
  • Posts: 835
Re: Nikon D5 - first impressions
« Reply #158 on: March 29, 2016, 09:24:51 »
https://www.youtube.com/watch?v=69U9RuJ5UnM

looks like somebody is not happy with the D5 :o :o :o

He might just be the new Ken Rockwell.
No way of telling what he did with the camera to reach these conclusions.
It seems to me all of his videos (or most, I just saw a couple) are geared to shock the viewer with unconventional opinions, no matter what the factual basis is. I'd rather just toss a coin to make decisions.
Simone Carlo Surace
suracephoto.com

richardHaw

  • Cute Panda from the East...
  • NG Supporter
  • **
  • Posts: 3133
  • Your lens loverboy
    • Classic Nikkor Maintenance and DIY
Re: Nikon D5 - first impressions
« Reply #159 on: March 29, 2016, 10:03:49 »
He might just be the new Ken Rockwell.
No way of telling what he did with the camera to reach these conclusions.
It seems to me all of his videos (or most, I just saw a couple) are geared to shock the viewer with unconventional opinions, no matter what the factual basis is. I'd rather just toss a coin to make decisions.
i just take it for entertainment value :o :o :o

stenrasmussen

  • Guest
Re: Nikon D5 - first impressions
« Reply #160 on: March 29, 2016, 10:14:00 »
The guy obviously never heard of the expression: "Better keep you mouth shut and let people think you're an idiot than to open it and remove any doubt".

Frank Fremerey

  • engineering art
  • NG Supporter
  • **
  • Posts: 12333
  • Bonn, Germany
Re: Nikon D5 - first impressions
« Reply #161 on: March 29, 2016, 10:29:46 »
To answer Andreas's question:

"Happy with a camera" does not mean technical data of any aspect of that tool need to be state of the art.

The discussion is all about the character of certain response curves of the sensor.

That value does not necessarily affect the user experience in any way.

Given Bill Claff's testing results the best sensors are Iq250 of Phase One, Sony a7r2, Nikon D600/610/750

Claff does not measure the whole package.

I wish for the D600 sensor in a better package, because this sensor delivers to my kind of photography but
the AF system and the sturdiness of the package leaves a lot to wish for.

On the other hand I never used the 9 fps my D3 can do.

Conclusion? I settle for the next 8xx body that will likely feature the Sony a7r2 Sensor in a D5 electronics and AF system
for 3500 Euros a pop. Meaning I can get two of these for the price of one D5. That way I can replace the D3 and the D600
with two exactly same bodies. Plus I get better sturdyness and better HiISO performance and significantly more resolution
in the studio.

If I was a sports shooter or into wildlife I would immediately get the D5.

It is the package that delivers to our photographic needs or not. A single technical aspect is ofter negligable.
You are out there. You and your camera. You can shoot or not shoot as you please. Discover the world, Your world. Show it to us. Or we might never see it.

Me: https://youpic.com/photographer/frankfremerey/

Frank Fremerey

  • engineering art
  • NG Supporter
  • **
  • Posts: 12333
  • Bonn, Germany
Re: Nikon D5 - first impressions
« Reply #162 on: March 29, 2016, 10:34:30 »
Concerning Youtube Stars: This is entertainment not science. If it upsets you it has created emotional value.

If it receives many clicks it is entertaining. It may be complete nonsense. Noone cares as long as it is funny.
You are out there. You and your camera. You can shoot or not shoot as you please. Discover the world, Your world. Show it to us. Or we might never see it.

Me: https://youpic.com/photographer/frankfremerey/

Erik Lund

  • Global Moderator
  • **
  • Posts: 6480
  • Copenhagen
    • ErikLund.com
Re: Nikon D5 - first impressions
« Reply #163 on: March 29, 2016, 10:44:34 »
Erik Lund

Ilkka Nissilä

  • NG Supporter
  • **
  • Posts: 1689
  • You ARE NikonGear
Re: Nikon D5 - first impressions
« Reply #164 on: March 29, 2016, 10:47:10 »
The  noise floor is variable depending on amplification that is applied, which depends on the ISO setting. Then there is the photon collection efficiency, which Nikon may have been able to improve (according to an interview by dpreview, Nikon improved the light gathering which I take to mean the QE). I would guess what has happened is increased QE and increased read noise at base ISO (due to faster ADC and perhaps a different amplifier design where high ISO is emphasized). Also the ability to do 4K recording may have required a different design which has more noise at low ISO.

As far as I know the 1D X was very successful for Canon whereas the D4/D4S were not so much for Nikon. The AF system may have been a major difference with the extended array of cross type points in the 1D X (now Nikon has that too). Nikon got some criticism from D3S users that the D4 was not much of an improvement at high ISO (few seemed to care that Nikon improved the low ISO DR in the D4 vs. D3S). So, Nikon may have decided to do what the market seemed to ask for:  optimize everything for high speed and high ISO. It is fitting that whatever they do, there are complaints online.  ;) Suddenly there are people who do care about the base ISO dynamic range in this type of a camera.  :o Well, perhaps they can now buy a D4s either used or discounted. For me I care more about the cross type AF points than a minor difference in low to mid ISO dynamic range but I do appreciate if the designers are able to make a fast camera also a good general purpose camera.

Personally, when they started increasing the base ISO dynamic range (with the D7000 and later the D800) I was sceptical of the usefulness and importance of it, but I quickly realized that the images genuinely looked better and could be edited in a more flexible way still retaining the quality. But I felt the D800 at ISO 6400 after correcting the white balance in low K lighting (candle light) produced a lot of noise in the blue channel and the images were IMO unusable (whereas the D800 at ISO 6400 in daylight was ok, since the blue channel didn't have to be amplified), so I felt there was a definite compromise.  The D750 is a bit better at high ISO than the D800 (or D810) with even underexposed ISO 6400 shots being ok after increasing the curve in raw converter, but the body is a  poor match for my hands (I find it very uncomfortable to use with gloves on, also it feels unbalanced with even mid sized lenses) so I look forward to what the D5 can do.