Author Topic: DXOMark tests Nikkor 400/2.8 FL VR  (Read 7367 times)

Jan Anne

  • Noob
  • Global Moderator
  • **
  • Posts: 2042
  • Holland
    • Me on Flickr
DXOMark tests Nikkor 400/2.8 FL VR
« on: July 31, 2015, 10:48:40 »
DxOMark tested the Nikkor 400mm f/2.8E FL ED VR on the D810 and compared to the model it replaced and its Canon competitor:
http://www.dxomark.com/Reviews/Nikon-AF-S-Nikkor-400mm-f2.8E-FL-ED-VR-lens-review-Revised-design



IQ wise the new FL model merely approaches the performance of the older model, it seems that the use of fluorite elements was purely chosen to save weight and not because it makes the 400 a better performing lens.

Maybe Tony can chip in, curious if he noticed any differences in IQ when using these big boys in the field.
Cheers,
Jan Anne

Ilkka Nissilä

  • NG Member
  • *
  • Posts: 1693
  • You ARE NikonGear
Re: DXOMark tests Nikkor 400/2.8 FL VR
« Reply #1 on: July 31, 2015, 12:37:34 »
Whenever I've seen images from the VR 400/2.8G in large sizes (e.g. NHM WPY gallery in London) the images have had striking sharpness (not to mention the vivid colour) compared to almost any other lens in the same show. It seems with the FL version Nikon solved the primary issue with that lens (i.e. weight and its distribution) and were able to maintain excellent image quality. I think with this class of high performance lenses, the image quality isn't going to change in large humps  as these lenses have a history of relatively frequent updates with the latest optics and technology. I wish Nikon put as much love into their shorter tele lenses (135mm please!).

Ilkka Nissilä

  • NG Member
  • *
  • Posts: 1693
  • You ARE NikonGear
Re: DXOMark tests Nikkor 400/2.8 FL VR
« Reply #2 on: July 31, 2015, 19:46:38 »
I notice that lenscore.org which test using a much higher resolution sensor, rate the 400/2.8E second highest of all the lenses they tested. 400/2.8G is 13th. This suggests that with the D810 the sensor resolution does not allow the new lens to show its best cards yet. Perhaps with a D7200 or a TC there is greater difference?

Lance B

  • NG Member
  • *
  • Posts: 111
  • You ARE NikonGear
Re: DXOMark tests Nikkor 400/2.8 FL VR
« Reply #3 on: August 21, 2015, 15:55:24 »
As Ilkka Nissilä has stated, LenScore has the 400E as the sharpest super telephoto that they have tested and they use a 200Mp sensor so the lens score is not sensor limited. Also, LenScore tests at 5 different camera to subject distances rather than the single camera to subject distance of all other test sites (that I know of) which gives a better indication of the lens's performance in real world conditions. This seems to be borne out when using TC's as I have found that the 400 seems not to be affected by distance when using any of the TC's, it seems as sharp in the distance as it does close up, whereas just about every other Nikon telephoto that I have used seems to have a observable drop in IQ at distance. My 300 f2.8 VRII and my 500 f4 VR both drop off IQ over about 20-30mts and it is noticeable. The 200-400 has been notorious for this issue. Maybe Nikon have paid attention to these criticisms of previous telephotos and have addressed this on the new 400. It may lose a point in sharpness at standard close up tests distances (however I dispute this - I think lens variation is probably the culprit), but excels at longer distances and hence why LenScore has ended up with much higher resolution figures for the new lens.

Having used the 400E for over 5 months and over 40,000 images, I can say it is the sharpest lens I have ever used. I own the 300 f2.8 VRII and the new 400 is sharper than that lens, but on DXO Mark the 300 gets the same score for sharpness. A good indication of this is that I get quite a bit more moire/false colour with the 400, not so much with any other lens.

What I will also say is that all the extreme tele's are bloody sharp and no-one could probably tell them apart as far as sharpness goes.

Erik Lund

  • Global Moderator
  • **
  • Posts: 6485
  • Copenhagen
    • ErikLund.com
Re: DXOMark tests Nikkor 400/2.8 FL VR
« Reply #4 on: August 21, 2015, 20:54:50 »
Please state what TC you're referring to ;)

I have fine IQ from the TC14E and TC20E III  on 300mm AFS 2.8

Sure these long lenses are outstanding  :)
Erik Lund

ColinM

  • NG Supporter
  • **
  • Posts: 1802
  • Herefordshire, UK
    • My Pictures
Re: DXOMark tests Nikkor 400/2.8 FL VR
« Reply #5 on: August 21, 2015, 23:34:56 »
Hi Lance, your real world experience is appreciated.

The NG forum doesn't currently allow us to see info for each member. If you could share any sites where you display your work (esp. When using the teles) that would really interest me.

Lance B

  • NG Member
  • *
  • Posts: 111
  • You ARE NikonGear
Re: DXOMark tests Nikkor 400/2.8 FL VR
« Reply #6 on: August 22, 2015, 08:47:28 »
Please state what TC you're referring to ;)

I have fine IQ from the TC14E and TC20E III  on 300mm AFS 2.8

Sure these long lenses are outstanding  :)

Hi Erik. I am referring to the 1.4x TCII and 1.4x TCIII, the 1.7x TCII and the 2x TCIII - I own all 3, sold off the 1.4x TCII when I got the 1.4x TCIII. The 1.4x TCIII is a little better at distance than the previous TCII, but the IQ still drops off a tad at distance. It has been talked about on many fora that most of the super teles seem to drop their IQ a tad over longer distances, especially the 200-400 f4. Under about 20-30mts, the 300 and 500 with TC's are excellent and sharp. In fact close in the 300 + 1.4x TCII or TCIII is almost indistinguishable from bare:

D800E + 300 f2.8 VRII + 1.4x TCII, f5.6,



crop of above



As you can see, there is no way you could ask for better than that!!

What I am noticing is that the 400 doesn't suffer image degredation over distance when compared to the 300 and 500.

Lance B

  • NG Member
  • *
  • Posts: 111
  • You ARE NikonGear
Re: DXOMark tests Nikkor 400/2.8 FL VR
« Reply #7 on: August 22, 2015, 08:48:16 »
Hi Lance, your real world experience is appreciated.

The NG forum doesn't currently allow us to see info for each member. If you could share any sites where you display your work (esp. When using the teles) that would really interest me.

No problem. My PBase site is:

http://www.pbase.com/lance_b

Erik Lund

  • Global Moderator
  • **
  • Posts: 6485
  • Copenhagen
    • ErikLund.com
Re: DXOMark tests Nikkor 400/2.8 FL VR
« Reply #8 on: August 24, 2015, 08:45:56 »
OK.

What my non scientific testing has led me to 'think' is that the build length of the TC is critical, that is; If it's not the correct physical length but longer or shorter, there will be issues with IQ...

Brian Caldwell should be able to contribute with further on this, and maybe Bjørn? ;)
It is a serious issue with the SpeedBooster design I remember

Yes I have seen the issues of 200-400mm AFS at distance, very obvious on the NG safari in 2009 I remember! I think we had at least 4 of those with us...

The other long telelenses with performance issues... ? must be due to air disturbance or lack of proper support / shutter speed or VR IMHO
Erik Lund

Lance B

  • NG Member
  • *
  • Posts: 111
  • You ARE NikonGear
Re: DXOMark tests Nikkor 400/2.8 FL VR
« Reply #9 on: August 24, 2015, 10:01:09 »
OK.

What my non scientific testing has led me to 'think' is that the build length of the TC is critical, that is; If it's not the correct physical length but longer or shorter, there will be issues with IQ...

Brian Caldwell should be able to contribute with further on this, and maybe Bjørn? ;)
It is a serious issue with the SpeedBooster design I remember

Yes I have seen the issues of 200-400mm AFS at distance, very obvious on the NG safari in 2009 I remember! I think we had at least 4 of those with us...

The other long telelenses with performance issues... ? must be due to air disturbance or lack of proper support / shutter speed or VR IMHO

No, I am sure it's nothing to do with air disturbance or lack of proper support/shutter speed or VR issues. I know how to use and test long lenses and from extensive use with the 1.4x TCII and compared to the new 1.4x TCIII, I can assure you it is better therefore ruling out any of the issues you refer to, ie it would manifest itself on the new TC as much as the previous TC if that were the case. I have also AF fine tuned the lenses and done critical testing to get the best from them with and without the TC's. I also know that the new 400 does not suffer from distance related IQ drop off when using any of the TC's, whereas the 300 does and the 500 did. It's not huge, but it is observable.

Erik Lund

  • Global Moderator
  • **
  • Posts: 6485
  • Copenhagen
    • ErikLund.com
Re: DXOMark tests Nikkor 400/2.8 FL VR
« Reply #10 on: August 24, 2015, 10:09:44 »
If your lens and camera body is calibrated by Nikon Service there should be no need to use any manual AF fine tuning. IMHO

Their lab equipment by far exceeds what is possible for DIY.

But then comes the TC... I don't know how much info is stored regarding AF calibration values in the TC and relayed onto the camera body.
Erik Lund

Ilkka Nissilä

  • NG Member
  • *
  • Posts: 1693
  • You ARE NikonGear
Re: DXOMark tests Nikkor 400/2.8 FL VR
« Reply #11 on: August 24, 2015, 12:17:04 »
I think autofocus accuracy may well contribute to some loss of sharpness at long distances. I adjust all my TC+lens+body combinations that have fast or moderately fast apertures before I use them. Another matter is simply the lens MTF. A higher MTF lens may tolerate a bit more atmospheric degradation before the image quality imperfections become visible in the image. Also a higher MTF lens may work better with a TC again leading to satisfactory image quality even in the presence of some degradation due to distance and atmosphere.

As for Nikon service calibrating the AF, it may be a good plan a) if you only have a few items to calibrate, or b) in some extreme cases where the camera or lens is way off typical, and c) if you have a service facility nearby. However I cannot give away all my camera bodies and TCs to service every time I add one lens to the kit (so that the bodies could be adjusted by Nikon to focus with the lens properly), or give all my lenses to be tested when a new camera is added and adjusted (not to mention if they adjust the lens, all the cameras will have to be adjusted as well, and may lead to problems when selling the lens for use with another camera). The equipment is purchased for my use, not for sitting at service. I prefer to adjust the combinations that I have myself with testing in my applications. it usually leads to satisfactory results or at least better than out of the box. I realize that service facilities have more adjustments that can be done than the user (and perhaps better techniques for measurement) but there is a trust issue with that approach. If they explain what they are doing, why they are doing it, and how they are doing it and I can monitor the work as it is being carried out and give my input to the process then ok, but I'm afraid this is more access than they care to give me.

Lance B

  • NG Member
  • *
  • Posts: 111
  • You ARE NikonGear
Re: DXOMark tests Nikkor 400/2.8 FL VR
« Reply #12 on: August 24, 2015, 12:34:18 »
If your lens and camera body is calibrated by Nikon Service there should be no need to use any manual AF fine tuning. IMHO

Their lab equipment by far exceeds what is possible for DIY.

But then comes the TC... I don't know how much info is stored regarding AF calibration values in the TC and relayed onto the camera body.

OK. Having had lenses "calibrated" by Nikon service, no, they don't have sophisticated test equipment to test for focus accuracy at distance. They showed me their test shots and they were of a building across the street! At close focus, yes they use test charts to accomplish this, but then so do I and I also use other real world objects as well. What they do is, they test at close distance on the test chart and then they check that AF Fine tune at distance by focusing on a building across the street. In the end, they weren't any better than I was at getting accurate AF Fine tune. 
TC's are all saved separately to the lens bare, ie, if you are have a 300 f2.8, then you have a calibration for the bare lens, but then you also have a separate calibrations for each TC attached, with a 1.4x TC attached and another for the 1.7C TC and another for the 2x TC.

Lance B

  • NG Member
  • *
  • Posts: 111
  • You ARE NikonGear
Re: DXOMark tests Nikkor 400/2.8 FL VR
« Reply #13 on: August 24, 2015, 13:06:44 »
I think autofocus accuracy may well contribute to some loss of sharpness at long distances. I adjust all my TC+lens+body combinations that have fast or moderately fast apertures before I use them. Another matter is simply the lens MTF. A higher MTF lens may tolerate a bit more atmospheric degradation before the image quality imperfections become visible in the image. Also a higher MTF lens may work better with a TC again leading to satisfactory image quality even in the presence of some degradation due to distance and atmosphere.

As for Nikon service calibrating the AF, it may be a good plan a) if you only have a few items to calibrate, or b) in some extreme cases where the camera or lens is way off typical, and c) if you have a service facility nearby. However I cannot give away all my camera bodies and TCs to service every time I add one lens to the kit (so that the bodies could be adjusted by Nikon to focus with the lens properly), or give all my lenses to be tested when a new camera is added and adjusted (not to mention if they adjust the lens, all the cameras will have to be adjusted as well, and may lead to problems when selling the lens for use with another camera). The equipment is purchased for my use, not for sitting at service. I prefer to adjust the combinations that I have myself with testing in my applications. it usually leads to satisfactory results or at least better than out of the box. I realize that service facilities have more adjustments that can be done than the user (and perhaps better techniques for measurement) but there is a trust issue with that approach. If they explain what they are doing, why they are doing it, and how they are doing it and I can monitor the work as it is being carried out and give my input to the process then ok, but I'm afraid this is more access than they care to give me.

OK. Having had lenses and camera calibrated by Nikon service, they filled me in on how they do it. If your lens is out of calibration, they can actually reset it in lens and there will be no need to have a calibration amount in your camera's AF Fine tune settings. So, if your lens needs +5 they can reset the lens to suit a +5 setting but actually show zero in the camera's AF Fine tune settings. In other words, there is a re-calibration that can be done in the lens. However, Nikon would rather you send your camera in as well for calibration as they will adjust your camera body with a reference lens so as to have a starting point and get your camera correctly set. If they do not get the camera set correctly, then they will forever be chasing their tail trying to get lenses set correctly for any other body that you may subsequently bring in for calibration. By this, I mean there is no point resetting your lens in the lens's memory to suit a particular camera as the next camera may require the lens to be something different. It is therefore best to get the camera set to a known reference point, using a reference lens that they use to achieve this. After they calibrate your camera to that starting point they will then effect calibration to your lenses.

Now as for the belief that TC's work better with fast lenses, I fully agree. However, my 300 f2.8 VRII is a fast lens and yes it does work better with TC's with less IQ drop, but it still suffers from distance related IQ drop as did my 500 f4 VR. Having talked to other users of Nikon super tele lenses and using TC's, they also notice a drop off in IQ at distance. No, it is not an AF Fine tune issue as we have all AF fine tuned our lenses.

Erik Lund

  • Global Moderator
  • **
  • Posts: 6485
  • Copenhagen
    • ErikLund.com
Re: DXOMark tests Nikkor 400/2.8 FL VR
« Reply #14 on: August 24, 2015, 13:37:39 »
...  (and perhaps better techniques for measurement) ...

It's a computer running a long series of tests directly on the lens while it's is actually focusing and hitting an/or missing target, storing the information in the lens CPU! There is no Perhaps in that sentence.

Nikon calibrate the body completely independent from the lens and vice versa.  No need to had over all of your gear...

As Lance states :)
Erik Lund