64 replies to this topic

[Alex_Mustard]

I have had the chance to test and shoot D3 and D300 today at a Nikon Pro Services day. Both are nice cameras.

I had the chance to play with the high ISO - I also have checked the preformance of low ISO (as both cameras have a base ISO of 200 - with low 1 and low 2 below that). And I have shot some test shots for corner sharpness with 14-24mm and my 17-35mm on the FX camera.

I was able to do test shots both inside and out. While it was not a thorough review (and the cameras were still in pre-release spec) I was able to investigate several of the features that I was particularly interested in.

The D3 also seemed particularly water resistant:


Alex

[3000psi]

WOW ... I am impressed how Nikon continues pay attention to ergonomics.

[Alex_Mustard]

I realise that there is plenty of interest in these cameras and I'll get something together later today. Here is a photo to keep you going!



ISO 6400 (cropped)

Alex

[loftus]

Talk about a teaser post!

[Alex_Mustard]

I must start all this by saying that this is not a formal review of the camera, nor did I attempt to make any objective comparisons with existing cameras. I had the chance to try both the forthcoming Nikon DSLRs today, D3 and D300, and I thought people would like to read my subjective thoughts.

I do not intend to buy either of these cameras in the short term. For me the advantages that the D3 offers over the D2X I already own are mostly not relevant enough to my photography to justify a new body and housing (these issues have been discussed many times on Wetpixel, so I won't repeat them here). This could change if I get specific commissions or end up with a housing for a hypothetical D3x that also took the D3.

I would also add that I think that the D300 also out guns the D2X. It is a big step on from the D200. DPReview said that the D300 is as much as a step over the D200 as the D200 was over the D100. I am inclined to agree. Although the FX chip D3 is getting all the attention, it was the D300 that really exceeded my expectations.

Some opening remarks:
There is no doubt that the D3 is clearly a better all-rounder than the D2X. The high ISO is very impressive. The AF on both cameras is very nice. The D300 felt slights slower than the D3, although because of the smaller sensor had better AF frame coverage. I felt there was a small improvement over the D2X, but not massive. The LCD screens are massive. It was shocking seeing them next to my D2X. The viewfinder on the D3 was not as amazing as I was expecting, and I didn't like the exposure meter being on the right. The D300 viewfinder is much nice than the D200. The live-view is cool! AF performance under live view is about the same as a standard digital compact camera. This could be very handy underwater. For those who shoot a lot, you will notice the camera also has a more Canon-like exposure - less under exposure than a normal Nikon! I really liked both cameras.

I'll make separate posts below for each feature I have comments on.

Alex

[Alex_Mustard]

I know that many people here are interested in high ISO performance of cameras. Personally, this is not relevant for my type of photography which is invariably at base ISO. But after seeing Charles Hood give a great presentation on deep, dark wreck photography at the Visions Conference at the weekend I do accept that it does have some point.

I shoot a D2X (as does Charles), which I think is one of the poorest high ISO cameras currently on sale! Charles says that he can get perfectly acceptable images for publication (indeed the shots he showed have already been widely published) with the D2X at ISO 800. The D2X is not perfect at ISO 800 - but if you are shooting wrecks in available light there is not so much fine detail to worry about. And a bit of noise adds atmosphere. Anyway, I am digressing.

In the digital camera hierarchy, the Canon 5D was widely accepted as the bench mark for high ISO performance. I have used this camera underwater shooting wrecks and I would say it is 2-3 stops better than my D2X. The bar was raised by the Canon 1D Mk3. This is considered by many to be 1-2 stops superior to the 5D (and on a different planet compared with the D2X, 3-5 stops better?). The reports I have seen (and consider reliable) about the D3 suggest it is at least as good and some say even 1 stop better than the 1D Mk3.

From shooting it today I would say that ISO 6400, 3200 and 1600 on the D3 all looked better than ISO 400 on my D2X (I don't feel qualified to comment on other cameras). But I would also add that ISO 200 on the D2X is certainly looked better than 1600, 3200 and 6400 on the D3. I am using phrases like "looked better" because this is a subjective test.

Alex

[Deep6]

Thank you for your review. With my diminishing visual acuity, I am interested in the “live view” function on the 3” LCD. Did you evaluate the hand held live view? Useful for u/w?

[Alex_Mustard]

I am going to post two series of ISO comparisons from the D3, I shot. Neither are perfect, as they were both impromptu without the control of studio conditions. Also consider that the D3 was running pre-release firmware and the final camera will be better.

The first scene was (annoyingly) mixed lighting - ambient tungsten and flash. And the balance between the two shifts (with more dominance of ambient lighting at higher ISOs). I have applied auto-levels to these crops to even out minor exposure differences (I shot on auto rather than manual, doh!). No sharpening applied. I was looking for a scene that would show some fine details. The crops are from the centre of the frame (which moves about a bit!).

Here is the whole scene at ISO 6400:


ISO 6400 (100% crop):


ISO 3200 (100% crop):


ISO 1600 (100% crop):


ISO 800 (100% crop):


ISO 400 (100% crop):


ISO 200 (100% crop):


Alex

[Dan Schwartz]

Dr. Mustard, and all:

First off, thanks for the hands-on evaluations, especially on the D300.

There's a (somewhat) counterintuitive technique for reducing chroma noise when shooting RAW at higher ISO's: Overexpose by 1/3 or even 1/2 EV.

If anyone wants the entire 18 page 1.5 MB PDF of the chapter below (which has the charts and sample photos), just send me a private message with your email address.

Let me quote from Getting Started with Camera Raw:How to make better pictures with Photoshop and Photoshop Elements:

[*SNIP* Removed by moderator for being off-topic]

[Alex_Mustard]

Thank you for your review. With my diminishing visual acuity, I am interested in the “live view” function on the 3” LCD. Did you evaluate the hand held live view? Useful for u/w?

The live-view is cool! AF performance under live view is about the same as a standard digital compact camera. Here is a shot I took with the live-view:



Alex

[Deep6]

The live-view is cool! AF performance under live view is about the same as a standard digital compact camera. Here is a shot I took with the live-view:



Alex

Humm, that would be in the tripod mode? How bad is the shutter lag, ala cp 5100? As best as I can figure out, in the hand held mode, you don't get AF, but then could one MF with the 3" view and w/o shutter lag?
TIA,
Bob

[Alex_Mustard]

That was shot while walking around - in hand held mode. I focused on a distant object, then switched to live view. Then aimed at the leave, pressed the AF button and then the shutter. There is not much lag - kind of like taking two photos.

But I don't really see it as a main shooting method - more an option when you don't want to/cannot look through the viewfinder.

[Alex_Mustard]

The other image quality factor that I was interested in was corner sharpness and vignetting - which has been a problem with some full frame digital cameras. For these tests I decided to shoot a wall! This was mainly to test corner sharpness and distortion, but it shows that the camera has little vignetting.

Here is a sequence of 4 frames to show the lack vignetting:



Alex

[Alex_Mustard]

And here are some 100% crops testing the corner sharpness with ultra-wide (weitwinkel) rectilinear lens (in this case my 17-35mm - about equivalent to a 12-24mm on my D2X). Corner sharpness with can be an issue with ultra-wides on FF sensors.

I also tried the new 14-24mm - wow. This lens seemed to show sharper corners - possibly because it has been designed for digital, unlike my old 17-35mm? It is also really, really wide. I was testing this with an architectural photographer who is very sensitive to distortion. His comment was that the 14-24mm had slight barrel distortion compared to the slight pincushion distortion of the 12-24mm and 17-35mm.

All taken at ISO 800. Auto exposure. Auto-levels and sharpen in photoshop. All corner frames are top left.

Corner sharpness was fine at apertures F8 and above (up to F22). Softness started to come in at F5 and was clearly visible when the lens was wide open (F2.8)

Centre of frame at F8:


Corner of frame at F13 (F16, F20, F22 very similar):


Corner of frame at F8:


I am presenting more F-stops at wide apertures as these show something.
Corner of frame at F5:


Corner of frame at F4:


Corner of frame at F2.8:


Alex

[Dan Schwartz]

Dr. Mustard, first off, thanks again for the hands-on report!

Anyway, I shoot brick walls too, as a cursory check for barrel & pincushion distortion. It's more difficult, though, to check for vignetting, since the target is already medium-dark red.

To check for vignetting, shoot a white wall with an 18% grey card taped to it in the middle, which you spot meter off of (i.e. put the card in zone V or VI). Then, any vignetting is easily measured in Photoshop by moving the cursor over the image and reading the changes in pixel values.

Hope this helps!

[Alex_Mustard]

Here is an example to show how wide the 14-24mm is on the D3:

[craig]

Alex, thanks for the information. I guess we can rule out shooting at 14mm f/2.8 now. ;-)

Dan, I see no point in polluting this thread with a long-winded, somewhat incorrect verbage on linear encoding. For those who want to be confused by the free intermixing of linear and logarithmic terms, they've been able to get that from Michael Reichman for a long time now. If you are reading threads on D3/D300 test results you probably don't need an explanation on how to "expose to the right". What is the point of discussing exposure in this thread anyway?

[Dan Schwartz]

It looks like the metering is being fooled by the spotlights in the ceiling, causing the exposure to be about 1/2 to 1 stop underexposed:



Dr. Mustard: Sorry if I'm being such a hard-ass today: A top-of-the-line dSLR (or any camera from Nikon, for that matter) should hand you better looking JPEG's than that when set to "Auto Everything" (exposure, WB, & such).

Here is an example to show how wide the 14-24mm is on the D3:
[see above]

[Alex_Mustard]

Dr. Mustard: Sorry if I'm being such a hard-ass today: A top-of-the-line dSLR (or any camera from Nikon, for that matter) should hand you better looking JPEG's than that when set to "Auto Everything" (exposure, WB, & such).

I didn't actually take this shot - so I don't know what settings were used. The architectural photographer (referred to above) did - on my card - while we were discussing distortion.

Alex

[EspenRekdal]

Hey Alex,

Thanks for the review so far. Can you elaborate on what this low 1 and low 2 is. Is it a Iso100 and Iso50 equivalent?

Thanks,
Espen

I had the chance to play with the high ISO - I also have checked the preformance of low ISO (as both cameras have a base ISO of 200 - with low 1 and low 2 below that). And I have shot some test shots for corner sharpness with 14-24mm and my 17-35mm on the FX camera.
Alex