Nikon D3X - errr..mm

[David Kilpatrick]

Very bad light conditions continued today. I have to complete the first report tonight, but there really wasn't any chance of seeing how the camera handles something as simple as a blue sky - and I'm especially interested in that, as the Sony handling of blue skies is pretty noisy. But, the 50mm f1.4 AF-S G lens arrived. I was expecting something big, like the Sigma, but it is not at all oversized despite the sonic motor. It's much smaller than Canon's f1.4, I am fairly sure without checking, and just a little larger than older designs. Even at a brief glance, it transforms the results (much as the Canon 50mm f1.8 transformed results on the 5D MkII). I may have a chance to get some kind of pictures tomorrow morning if the cloud shifts, but it's very frustrating when light levels outdoors all day are lower than in my office with its one energy-saving bulb!

The guys on the dPreview Nikon D3/etc are making a bit of a meal of my comments, maybe unaware that I have often come out with similar comments about the Alpha system - a couple of them are suggesting that I find no problems with Sony, but criticise other makes. That is not so as the main article pages on this site demonstrate, more than a few are devoted to highlighting issues with the Alpha models, lenses etc. If I find something which is worth making known to users, of any system, I'll try to do so. Photoworld magazine includes references to shortcomings, limitations, workrounds, third party solutions etc in a way which none of the other 'user club' magazines does. We continue to have no connection at all with Sony financially or in any other way (so far all test items since September 2007 have been bought, not borrowed for review) and it may be that this approach closes the door, but if that's the case, so be it.

And just for the ref - one poster there reckons using P mode in low light is not very professional. Well, using M mode was not an option; using A, I would have been obliged to set full aperture of f3.5 and let the lens force this down to f5.6 when zoomed, and of course the shutter speed would always have been exactly the same speed as P mode, which also selects full aperture. Using S mode, a speed of 1/125 might have been possible, but the lens would then have been wide open and only given correct exposure at 24mm, as the camera does not over-ride the S setting once full aperture is reached and the light is too low. In fact, straight old Program mode is the best choice under these conditions as it will take into account focal length and attempt to keep the shutter speed appropriately short, while selecting a slightly stopped down aperture should some miracle of unexpected light hit the subject. My choice of ISO 800 was based on tests earlier which showed 1600 to be too noisy for my intended use, but a -1 over-ride was set knowing that nearly all the subjects would have the active metering (using matrix) on a relatively dark area, in conditions where overexposure would be very undesirable. And, based on my experience with the 16-85mm VR Nikon, I was confident that the 24-120mm would be sharp wide open and have highly efficient stabilisation; I have now learned (and did not know before) that this lens has a variable reputation and many former users consider it one of the worst in the range.

Program mode is entirely compatible with informed use; Aperture priority would have given identical results; Shutter priority would have resulted in some images being underexposed.

David

David

[read the manual]

"I am quite aware of the functions of the D3X and have used the D3 and D700 before, and read the manuals (cover to cover, every page, followed by extensive back reference later)."

Thats great David, so next time you use the tilt and shift you won't need to "Nikon: eyepiece blind closed had to be opened between shots to allow adjustments via the viewfinder, no other easy way to change settings. The top LCD display was facing in a direction which gave me a headache bending round to try to see it, in shadow from the studio lights, too small - etc." just press the the info button instead as you back reference to page 14.

"Even the rear screen reads sideways, the Sony automatically rotates the info." nice feature for the Sony.

[pakodominguez]

An issue I have, specially with the 17~35 f2.8-4 is that it takes a little to long to fix the focus: some times it looks like the camera got it and, when I re-compose, the focus go away. Since most of the time I'm working in the dark, I don't always realize my focus was gone to the infinite and beyond -never better said!

Probably the issue with the lens David is using (other than a moderately poor IQ) is that is a slow focuser and, when he re-compose the scene, the camera continuous focusing... why don't you try to get an "old" Nikon 35~105 f3.5-4.5? a nice and compact lens, I'm sure you will get nice results even on a 24MP camera

A nice feature on the Canon cameras is that, when chimping, you can see where the camera focused (well, if you use the central AF sensor doesn't really helps)

Regards

[David Kilpatrick]

Don - Interesting - I spend nearly all my time in A mode normally, to control depth of field of ensure I'm stopped down enough to avoid wider aperture problems, and not so far to risk diffraction. Generally this just means setting f8

It's surprising how slow a shutter speed can produce acceptable results if the focus is correct, even with a pretty basic camera and lens. Here is another set of pipers, caught in almost non-existent late evening light - 1/25th at f5.6 (wide open) on the Canon EOS 1000D with 18-55mm IS lens:


http://www.pbase.com/davidkilpatrick/image/107156980 for full size

OK, there's no way the noise or image sharpness from the Canon is acceptable at ISO 1600 even when processed from raw with ACR to sharpening 25, NR 25, CR 50. The idea that all Canon cameras have noiseless high ISO rests on a few models and the 1000D is not one of them. But even this porridgy result is almost acceptable because the faces are reasonably sharp. I'm pretty sure it was in program mode too and that whatever the AF system was offering, it was either set to centre sensor focus (my normal preference) or factory default.

Ref the manual and D3X - agreed, I should have used the rear screen display, but after a couple of weeks using the 5D MkII and working alongside the Sony, Quick Control and Quick Navi both use the rear screen in a different way - both automatically responding to control use, and being a control in itself. With the D3 I made hardly any use of the rear screen and mainly relied on the viewfinder, and with the D3X it probably did not even occur to do anything other than continue using the finder display when I changed from flash shooting to tungsten A exposures. I would have remembered pretty quickly if the camera was just above 'relaxed eye-level' on a tripod. This is the one thing which has been made SO much easier with the adoption of rear screen info displays.

David

[David Kilpatrick]

And maybe I was being over cautious. The Nikon JPEGs told me that anything over ISO 800 was really fairly poor, and I was not all that impressed with 800 either. But while writing stuff up tonight, I messed around and made some exposures at 6400 with the 50mm f1.4 - just sitting at my desk, aiming at a guitar on the wall (yes, it has seven strings, it's Russian, no Photoshop faking). Terrible light. I decided to check the result using RAWDeveloper, because I ran some of the ISO 800 and 1600 raws through this and the difference was amazing - MUCH finer than the in-camera JPEGs.


http://www.pbase.com/davidkilpatrick/image/107161372

Screen shot showing no sharpening but default NR, 6400 at 100 per cent view. Capture NX2 is not doing as well as this, it produces something more like the JPEGs. ACR when it comes will probably not be as fine (RAWDeveloper also makes A900 high ISO stuff much better). This is the problem of trying to use a camera before all the raw conversion tools have arrived.

I conclude, for the moment, that the D3X may work almost as well as the D3 in low light/high ISO - from raw, and with future raw converters.

David

[InTheSky]

...

This comes back to the point Frank made...which is formation walking/marching can take a slower shutter speed because you can catch everyone in the same relative state of rest (minimal motion in this case). Yeah, if the Nikon would have actually hit the Klauses at 1/45 more parts of more of the red fellows would have been in focus.

...

-Sonolta

[/quote]

Ho my god Sonolta you are referring to me !

[UrsaMajor]

David,

In an earlier message you mentioned that you had heard that the Nikon 24-120mm was not a good lens. On the chance that you were referring to another source, I'll mention the comments about that lens made by Mike Johnson on his web site, The Online Photographer, on 28 November. Take a look at:

http://theonlinephotographer.typepad.co ... 120mm.html

In that column he refers to the Nikon 24-120mm as literally a piece of fecal matter, with which he was unable to get decent pictures most of the time. He was using it on a D700, with which he was very satisfied when he used other lenses. With both of you having had problems with that lens, and on different Nikon models, it certainly sounds as if that particular lens suffers from either design issues or significant quality control problems.

With best wishes,
- Tom -

[David Kilpatrick]

B*gger! I had not seen that report - it is more explicit in firmly describing the lens's problems than I would have been. I agree on all counts, including the not getting sharp in any way at 120mm, and the failure to be able to focus at all (this had me baffled) plus the apparent non-effect of the VR.

I have been looking at my useless Santa images and conclude that something very odd has happened. I think the VR mechanism has 'corrected' my slight panning action as if it was camera shake, resulting in some odd double imaging effects on the unfocused subject, while keeping the background sharp.

David

[Greg Beetham]

So, it's looking like it's not just that individual lens that may have suffered some misadventure, more like the lens model itself that isn't so hot, one wonders why they allowed "that" lens too be tested on the D3x seeing as there would be a great chance of it not performing very well, or worse, casting suspicion on the camera itself?? I mean surely Nikon would know that model lens has issues by now....OTOH it looks like the camera will perform well when there are a few good lenses to choose from, going by that 50mm shot.
Greg

[01af]

B*gger! I had not seen that report ...

Huh!? I pointed that article out to you five days ago, and you hadn't seen it?

-- Olaf

[Greg Beetham]

I forgot to go back and read it as well Olaf, sorry....he certainly hasn't much good too say about it that's for sure.
Greg

[Javelin]

I had read somewhere. somewhere being that place in memory you put things that need to be substantiated. that the lens based VR systems don't cope well with panning motions where sony's SSs can detect panning instantly and lessen the effects of SSS on one of the axis to compensate for the intended movement. the problem is there isn't a good technical descriptions of these systems around so you can learn exactly whats happening with. them but reading David's comments makes me think this VR vs SSS is true.

[Javelin]

I think the VR mechanism has 'corrected' my slight panning action as if it was camera shake, resulting in some odd double imaging effects on the unfocused subject

I can only go by what he said

[David Kilpatrick]

Different sort of panning, of course I did not mean for a moment that I was executing a follow-through action pan. Just meant the camera may have been moved in the general direction of keeping the subjects framed.

The 24-120mm does not use the right sort of VR for panning. It's got the old original type, no idea why for a relatively recent lens. The 16-85mm DX equivalent has intelligent VR which can cope with deliberate camera movements. So does the 24-70mm. This is just one of many disadvantages of the 24-120mm I was not aware of.

David

[01af]

Today, Mike Johnston has published a new article on his blog, comparing his impressions from the Sony A900, Nikon D3 & D700, and Canon EOS 5D Mk II (however, no D3X). See here:

http://theonlinephotographer.typepad.co ... kon-v.html" onclick="window.open(this.href);return false;

His bottom line: He likes the A900 and the D700 best, both for different reasons. He does not like the EOS 5D Mk II so much because it's too much of 'jack of all trades, master of none'---which however might be considered also a strength, depending on your point of view. Anyway, he feels the Canon produces a 'digital look' more than the others do.

-- Olaf